They had faced a gruelling few days on social media, caught up in allegations that the organisation had facilitated Israeli nationals to cross the Malaysian border. Founder Balaji Srinivasan had dismissed the rumours as fake news, taking to X with a video announcing the pause of investment until the receipt of an official MOU from the Malaysian government and questioning whether Malaysia would be a good place to keep driving investment at all.

Share

The police conducted an inspection of all 266 of the international community’s passports, lining up NS community members in the lobby. Ultimately all had been found in order, and the community took the opportunity to take a group photo with passports in hand.

Yes, there were still queries around some the business licenses on one of the coworking units, and the wording of a sign, but they had time to solve it. Some of the residents of Network School had been (and still were being) subjected to online hate and violent threats.

But the morning of the 21st it seemed like the media storm had passed.

Then the Iskandar Puteri City Council, the jurisdiction covering Forest City, issued a notice on X that the business license of Network School would be revoked from the following day. Enforcement officers arrived on 22 July, sealed the entrance to the coworking space (a separate building from the main campus), and affixed a closure notice. Johor’s Menteri Besar, Onn Hafiz Ghazi, said that no investor, company or organisation could be placed above the rule of law.

Following a meeting of the whole community the decision was set – they would leave in search of a new location.

Within hours, residents were packing bags to leave Forest City. During a short videocall with a friend that was there at the time, they showed me the queue of taxis lining up to take them to their next destination. Already, they told me, some had made their way out to a second node in Indonesia despite reports that the facilities and WiFi were basic. Others made their way to Vietnam and other surrounding countries. The general consensus seemed to be, “Let’s get out, fast, but somehow stick together.” And so, the residents of NS scattered, cast into the world ready to find their next base.

The day after the closure notice went up, Balaji was in Kazakhstan signing a five-year memorandum of understanding with the Ministry of Artificial Intelligence and Digital Development. He promised a new campus that would be “a haven for global techno-optimism, with expedited visas, streamlined redomiciliation, and active recruitment of talent.” A pathway for NS’ next iteration, however unclear on specifics, was set.

Options, options

My timeline in the weeks since, posts and messages authored by onlookers like myself, has been one long argument about what should have been done differently. Everyone is proposing a different route, and most of the suggestions have already been tried somewhere else with mixed results.

Many supported the approach Balaji has now acted on, which is that you get the paperwork before you arrive and then set up the community. A five year MOU with Kazakhstan is fundamentally a document that says two parties would like to work together covering hackathons, joint events and possibly a Network State conference before the end of the year.

What makes it meaningful is the fact that a named ministry has publicly attached itself to their existence, which could raise the political cost of getting rid of them later…or make them a target for an opposing political party in the future – but regardless, to some extent NS has now entered the political landscape with an agreement.

The second idea floating about is that a memorandum is nowhere near enough, calling for legal entrenchment, which is where Próspera keeps getting invoked. The ZEDE framework in Honduras was written into the constitution in 2013, complete with its own courts, its own regulatory regime and a fifty-year stability guarantee designed specifically so that a future government could not change its mind. A government changed its mind about ZEDEs in 2022, so Próspera filed for arbitration seeking around eleven billion dollars. Four years later it is still operating and still litigating. So the entrenchment worked, in the sense that Próspera is still there.

The third stance I’ve seen come up is that the whole framing is wrong and portability was always the point. Network School already had a second node in Batam, Indonesia, which is where some residents went first, and many of the rest of the community has now migrated to different countries in groups.

On this reading nothing failed at all, since a node closed and the network routed around it, still intact but scattered. It is a bit reductive though, not accounting for the disruption the eviction caused for the NS residents, many of which are startup founders and freelancers that struggled to keep continuity for customers and clients around their already heavy workload, the uncertainty of their location now adding to an already uncertain worklife.

Local connections

A couple of weeks ago I was browsing X when I noticed a Logos space was underway, with Sterlin Lujan, founder of Polis Labs and Infinita founder, Niklas Anzinger.

While the main subject of their conversation was “playing the infinite game” Network School came up along the way – as I mentioned above, the events around Network School has sparked conversations around how Prospera approached government pressure and Niklas was there throughout the charter city’s most recent clash.

Niklas spoke about the different approaches the two projects had taken to developing within their respective countries and highlighted Network School’s quick turn around from Malaysia to Kazakhstan. Sterlin, however, pushed back:

“I do think that more could have been done on the ground in terms of local community buy in to help prevent some of the ire that was manifested in the local community [...] I think a lot of the locals in Malaysia felt snubbed that he wasn’t regularly talking about them and they didn’t feel included.”

He spoke about when he visited the campus as part of the first cohort and how, when he spoke to locals in the surrounding areas they viewed forest city as “cursed land” - “as a result of this they really were skeptical of any kind of foreign investment that flowed into Forest City, and the result of that is automatic distrust. There needed to be more of a strategy around how can we gain the local buy in because they’re not going to trust us out of the gate.”

When I first visited the school, I had noticed that there wasn’t much interaction with local Malaysians. Although many of the attendees had made an effort to visit the surrounding area, learn basic Malay and play volleyball or football together despite the language barrier, it was very easy to ignore the outside world completely. Perhaps that was part of the point – Network School was marketed as a place to “lock in” without distractions, allowing for startup founders to go from their bedroom to their desk and back again if they wanted to, with some encouragement to balance it out with a gym session.

The second time I went though, more recently, the approach had changed somewhat. There were Malaysians in the core team, and more the exam to join. There were events to collaborate with local tech startups, and the community newsletter (private to longtermers) showcased local tech-forward initiatives and innovations. There was movement, but perhaps too little too late.

The dispersed network

I’ve thought a lot this past month about how the events around Network School Malaysia unfolded, and how the next iteration of it might look.

I asked a community member who was there though as the events rolled out, “what do you think we can learn from this, has it changed how you approach the space?”

“No,” they said, “everything was entirely predictable. We learned nothing from any of this other than we are complete [idiots] [...] Johor was starting to work, it could have been a massive success story in two years but instead f***** it.”

They didn’t expand further but presumably weren’t referring to the idea of a network state itself, later sending photos from the new community they had joined.

In one sense, the events around the end of Network School Malaysia have proven the network state thesis. The connections made within the walls of Marina Hotel, Forest City, have outlived the physical location, instead using their digital means to expand, adapt, and come back together in places that are a little less remote. I know of a few that for now at least are burnt out by the whole experience, others still that are preparing to leave their home-towns to join one of the latest iterations.

Through scattered messages and voice notes, peppered with the occasional photograph, the month following has been one of scattered regrouping. Of the offshoots I’m aware of, there are now groups in Vietnam, Indonesia, Sri Lanka and India, as well as the new NS gearing up to welcome residents from 1 September.

Will all these new offshoots strive for government recognition? Probably not. Some just want to find a place where they can stay as part of a community and build. In that way they seem like more of an evolution of hacker houses and live in incubators, which might be slightly more palatable, even embraced by a host country.

Share

As for those that do want to reach the end goal of government recognition, it’s a long game which was never going to get resolved within a matter of years. NS’ MOU with Kazakhstan is for five years, perhaps after which it evolves into something more permanent…or it acts as a test case for other jurisdictions to reference.

The month since has demonstrated something the movement is now good at, which is quickly adapting. Two hundred and sixty-six people were told to go and were gone within days, scattered across four countries, with a ministry in Kazakhstan signed up before the closure notice on the coworking door had faded in the sun.

But Johor did not close Network School over a business licence. It closed it because several hundred people had lived in Forest City for the better part of a year without building the support online and in person from people in the country.

In his discussion with Sterlin, Niklas said that this is harder than it seems - that local relations can be good but “you won’t ever completely avoid that potential for backlash,” which often, he said, comes from politicians in far off capitals. “You do need a deliberate strategy to survive.”

The new campus opens on 1 September, with better paperwork than Malaysia and a named ministry attached. The next months will show the strategy around building local support.

First, there was, obviously, the title, which, having called this Substack Utopia in Beta, seemed pretty aligned, but then, when I went into it more, I realised the book could potentially shed a whole new light on moonshots, techno-optimism, utopian thinking and, possibly, network states.

(Plus, the introduction closes out with “I have no answers, only a restless machine for asking questions,” which is sometimes what I regard this Substack to be.)

Now it is worth mentioning that Joad hadn’t heard of network states until I approached him for an interview. His book centres on the tradition of literary thought around utopias rather than the practical implementation of innovative thought to building societies. He does, however, go into a few real intentional communities which apply utopian theory, such as the 1968-founded Auroville in Tamil Nandu, India and all the way back to Vasco de Quiroga’s attempt to build communities in Mexico based on Thomas More’s utopia in the 1500s.

As you’ll read below, he sees the concept of “utopia” acting more as a mental tool for critical thinking, which is somewhat undermined when thinking about practicalities. In fact, being unattainable is often part of the point - allowing people space to take an idea for a society as far as it can possibly go.

Share

In this conversation we trace the utopian genre from Plato and Thomas More through Margaret Cavendish and Sarah Scott to Sun Ra, and test it against the thing I cover every day, network states and the Silicon Valley builders trying to make the unattainable real. We talk about paradox, freedom, why these books leave the individual out, and what happens when you try to lift a piece of a utopia into the real world.

What we cover:

• Utopia as a machine for questions rather than a blueprint, a tool for thinking drawn from a tradition of literary texts and real-world experiments that only sometimes overlap

• Whether network states belong to the utopian genre, tested against founding moments, founding myths, law, property, equality, education and shared culture, and why Wren would trace them back to Plato and Ecbatana

• Paradox at the heart of it, with Plato building an ideal state through a fiction that then bans fiction and exiles the poets

• Why attainability barely matters, with Margaret Cavendish’s deliberately impossible Blazing World set against Sarah Scott’s plausible Millennium Hall.

• Moving the word utopian away from impractical and unrealistic, toward dismantling the values that underpin the society we live in

• Freedom as endless individual choice today versus the older utopian freedom of internal discipline, with William Morris and Ursula Le Guin as the exceptions

• Why utopias dissolve the individual into the collective.

• Utopia as a path to incrementalism, the role of moonshots, and Silicon Valley as the most active utopian builder of our era, and whether network states, seasteading and Mars projects extend the tradition.

Joad Raymond Wren
This is going to be very interesting because we come from completely different worlds and this is probably going to need to be much more of a collaboration. I heard about Network States this morning.

Isabelle Castro
Oh, really?

Joad Raymond Wren
That could have been the final chapter of the book after Shy Radicals. So I think there’s. There’s very much a dialogue between what you do and what I do. It’s fascinating.

I can see how there’s a kind of trajectory there. And if I went a little bit more into the 2020s, it makes a great deal of sense and there are loads of questions to be asked about.

Isabelle Castro
So what are some of the questions that come to your mind having heard about Network States this morning? What were the things that kind of jumped out to you?

Joad Raymond Wren
So the first thing I thought was about to what extent does this fit into the genre based model of utopian thinking? Which is to say that if I were to look at the project or the projects that follow from that book, SRI Lans book as a whole, to what extent would they fit the criteria that are established by the tradition of writing about utopias?

Do they have a founding moment? Yes. Do they have a kind of a founding myth? Possibly. And I think it would come down to questions about ideology and the perfectibility of society through AI, through online communities. But then are there answers provided to the kind of standard issues of the organization of the law, international relations, the questions of private property, various questions of equality, gender equality, where does education fit into this state? And to what extent is there a shared culture and what is the role of that culture in upholding the law, in furnishing some kind of political stability and continuity?

Share

I don’t have any answers to those questions. But immediately I was thinking, okay, to what extent is this a recognizable project? And it would probably be really annoying for me to do, but at the same time it would be, I think, quite interesting to have that perspective and say, okay, we need to go all the way back to Ecbatana in the 7th century BC and to look at Plato and say, okay, is this a little bit more familiar than we might think?

Isabelle Castro
Yeah, a lot of the founders within the space, they often look to Plato at the idea of the philosopher king so I really enjoyed reading that part of your book in particular.

What I liked about the book is that you said yourself that you’re kind of like a machine for questions rather than kind of a prescriptive blueprint approach. In your mind, what does Utopia and the idea of utopia signify for those who have questions?

Joad Raymond Wren
I guess I would say that utopia is a tool for thinking that is extracted from a tradition of literary texts alongside which runs a series of projects of trying to actually establish utopias in the real world which do and don’t overlap. I think sometimes it’s quite easy to use the word utopia and to assume from that it means a kind of a fairly systematic good place to live that we can establish within our society by putting certain kinds of walls, whether they’re physical or financial or political, around us and within that, those confines, within that island to build an improvement upon society. Now sometimes that overlaps and it directly connects with thinking about Plato and more and so on.

And sometimes it’s a tenuous association, such as in Auroville, which I wrote about in the last chapter, which I know it’s fascinating, but it’s not strictly speaking within the tradition of thinking of utopia that I’m writing about, but at the same time it’s a very serious and long standing and not entirely unsuccessful attempt to establish an intentional community.

I guess that was the other thing that I found very interesting about your Substack was that I was kind of unaware about the revival of interest in intentional communities in the 2000s, which it seems to be very much a thing of the moment. When I think about [intentional communities] I think about farming communities in the Basque region and Auroville and the commune I was born into and places like that. Not these kind of interesting hybrid, real world, virtual world phenomena.

Sorry, I’m getting sidetracked. For me, I guess utopia is a way of reflecting upon the world in which we live and as importantly, the way we think about the world in which we live.

So, at the core of that way of thinking is the paradox. And the paradox is really important. Not because it’s a clever fictional or literary device, but because it brings together competing sets of presuppositions that are in some ways incompatible. That makes us think about their incompatibility. So in terms of what Plato does, Plato’s foundation stone for the Republic is that people have to do what they’re good at. The optimal society has to optimize all of its resources. And it does that by making sure that people do what they’re best at, and they only do what they’re best at.

In order to affect that, language needs to be used in a particular way. So it needs to be used clearly and rationally and unambiguously. And it can only refer, words can only refer coherently to one thing. You can’t have ambiguity in language if you’re trying to analyze and understand the ideal organization of society. You put those two things together and you have to ban poets. So, you know, the other famous thing about, apart from philosopher kings is the abolition of poets. They’re kicked out of society because they use metaphors. They use language in a way that is not true, not literally true, but also they pretend to be people other than they are. They pretend to be gods, they pretend to be carpenters, they pretend to be things in their poems other than poets.

Share

However, the Republic is a fictional dialogue set in the year in which Plato would have been about 10. Those people probably weren’t all in the same room at the same time. There are arguably one or two anachronisms in relation to Athens at that time. And the whole thing is a dialogue between a fictionalized version of Plato’s actual teacher and a number of other characters. And the premise of the conversation is that they are arguing, they’re debating whether happy people are good and good people are happy. In order to do that, Socrates says, okay, I know, because the human mind is a small thing, and it’s hard to open it up. What we should do is we should talk about this term in terms of the city.

He creates analogy between the mind and the city and says, let’s not talk in terms of parts of the brain, but in terms of the organization of an ideal commonwealth. So the whole thing is a metaphor. So in order to answer his question about whether happy people are good and whether good people are happy, he creates a fictional state using a metaphor. But the premise of the answer to that question is the abolition of fiction and metaphors. The whole thing is a magnificent and beautiful paradox. And so that was a long answer. And the same is true of more. And the same is more or less true of most of the books that I write about. But what it means is that particularly Plato, and particularly More, that they’re saying to us, okay, this is how we think about things.

Why do we think about things. Things in this way? What is. What are the limits that we impose upon ourselves by saying to ourselves that certain things are beyond consideration, they’re beyond belief, they’re against belief paradoxa. And therefore they’re using the creation of an ideal commonwealth through paradox in order to challenge. Not to say it’s would better if we abolish private property, though that is frequently a premise. But to say, okay, if we can’t accept the abolition of private property, if we can’t accept the things that appear to follow from accepting that premise, how. How are we limiting ourselves? Is there practical utility to thinking impractically? To me, it’s a kind of. It’s a kind of philosophical exercise or a political exercise conducted through fiction, which means that what matters is not the creation of imaginary world which you may or may not like and you may not or may not think has merits and virtues, but a kind of philosophical argument conducted through fiction that is able, through fiction, to make a philosophical argument. And the same with politics. So it’s a kind of political argument conducted through a kind of fiction that is able to make a political point that you can’t do through ordinary political writing.

Now, that seems to me to be the real merit of utopia is not just the imagining of a better or a perfect world, but the engagement of our own minds in reflecting upon our own values and the relationship with our actions and the political structures that we endorse around us.

Isabelle Castro
How does attainability or the belief that it is attainable factor into how well that kind of mental exercise works in your mind?

Joad Raymond Wren
You see, I’m not sure it does now? I mean, I guess it depends on who you’re writing about. I mean, I guess if you think about the feminist utopias. Okay, so one way of thinking about that, I guess is Margaret Cavendish and Sarah Scott. So Margaret Cavendish, writing in 1666 in her blazing world, this is one of the few utopias that actually is quite an interesting story. Usually the story is a bit lame, but she imagines a world in which a highborn woman is abducted by a lower born man who can therefore never win her hand by wooing. Her abductor takes her to the North Pole. For whatever reason, all of the men on the ship die. She survives and she discovers at the apex of our world or her world, not our world, her world, another world. And she crosses into that world.

So it’s a real Philip Pullman kind of multiple universes connected by invisible bridges of poles kind of scenario. She meets the emperor of the world, is immediately made empress and invited to rule the world. And she decides that she wants to invite somebody to act as a kind of literary secretary in Emanuel. So she invites this woman called Margaret Cavendish. She has this great. She says, I know, I’ll have Descartes, I’ll have Gassandi, I’ll have Hobbes. And the spirit who’s helping her says, no, they’re all men. They’re much too vain. You don’t want them. Margaret Cavendish. So Margaret Cavendish is invited as a character into her own fiction, typical utopian kind of device. And she decides that she wants to rule her own world. And so she proposes to the same spirit various worlds that she might take over.

And he says too problematic, because that would involve conquest. Conquest is an important issue for utopias, or at least military self defense. And so he says to her, well, just why don’t you make one up? And so she creates a world in her imagination. What we might understand about the actual existence of that world is slightly complicated by the fact that she appears to believe that all matter is sentient. She doesn’t believe in a separation between matter and spirit. So in a way this fictional world she creates has some kind of attenuated being in reality. And so she governs this world and she devises the perfect society. But of course, the very cost of Margaret Cavendish, both fictional and real. Margaret Cavendish is that the world that they create is fictional. It cannot exist. It exemplifies the most orderly form of knowledge.

Margaret Cavendish, the author, is very engaged with the Royal Society. She’s the first woman ever to attend a Royal Society meeting. Which is the year after she publishes this book. She’s much mocked for so doing. She doesn’t exactly break the glass ceiling because women aren’t admitted to be members of the Royal Society until 1945. But in 1667, she does go through the doors and watch an experiment involving an air pump demonstrating the weight of air. So she’s. She’s very much engaged with the nature of what we would call science, what was then called natural philosophy, and that this informs her world. But the very cost of the kinds of truth that are available in her world is they don’t exist. So unattainability is precisely the point of that book.

Share

But in a sense, it’s a challenge to kind of patriarchal values and an assertion of certain views about how science should be conducted through that unattainability. Flash forward about 100 years to Sarah Scott’s Millennium hall in 1762, which is another kind of. This is a separatist utopia in which a number of women come together, having had awful experiences in their lives, mainly to do with marriage and inheritance, and using the money that they gain through inheritance and marriage, establish a world that’s semi segregated from the outside world. It’s encountered by a relative of one of the women who happens to have a coaching accident as he’s passing their estate. Coaching accident. Same as a shipwreck, you know, same as a spaceship crash. And other kinds of utopias. The, The. The world that they’ve created is. Is one in which men only act as servants.

There is a school for. For local boys within the confines of their grounds, but boys are only educated to do kind of manual work, whereas women are educated for more. More intellectual purposes. This society is clearly, in some respects attainable, and it is equally a critique of patriarchy and particularly the marriage system, also a critique of the failing. Failing, in Christian terms, morals of the society around them. But in a sense, its attainability is. It is. It’s plausible that it could be attainable. It’s not fantastical, and in a way, it’s paradoxical in another sense, I think, because there are a couple of hints in the book that the money that provides for creating this whole is derived from the slave trade. And in future instantiations of separatist utopias, I think there’s some reflection on the relationship between patriarchy and slavery and so on.

But if you look at the contrast between those two books, one is the whole point is that it’s unattainable. The other it is attainable, but actually what they’re doing is engaging in some kind of ideological critique and some kind of challenge to how we take on board and internalize the social and gender hierarchies of patriarchy. So that was an insanely long answer.

Isabelle Castro
No, I like. I like long answers. I really like long answers. And it’s given a lot of food for thought. I wonder, kind of. I appreciate that the book kind of looks at kind of historical, literary pieces and all of that, thinking about kind of now what in your mind would be the critique that certain utopias would be focused on? I don’t know whether in your research you had a chance to look into whats going on now or whether you’ve thought about it.

Joad Raymond Wren
But the book kind of ends around 2017, 2018, and. I guess subsequent to that, I’ve been thinking about musical structures and utopia. And the book ends with some brief reflections on Sun Ra. But there have been a number of recent albums I still listen to albums about which have utopia in the title or are about utopias. I mean, Gueno’s album the d’ Olave is actually based upon a Welsh utopia. And I guess that instantiation takes us away from the immediate political concerns of the inequalities of modern society and the threat to democracy that plutocracy represents and so on and so forth, towards quite abstract structures. And I do think that is a. One of the threads that we see running through the second half of the 20th century. It’s a kind of retreat from immediate political engagement towards thinking about aesthetics.

I don’t think anybody would want to hear my reflections upon the 2020s, to be honest. I have to say that again, looking at your substack this morning, I think there is a whole world there that I have kind of failed to consider.

Isabelle Castro
I think it’s one of those worlds that…Because I’m primarily a tech journalist, I do engage with it on a daily basis. And I. Yeah, I can see why there is a retreat from the political engagement, because it just seems. It’s so immediate that it does. It hits quite hard. But I. In a way, I see kind of network states as a response to that, as a practical approach to trying to solve some of the issues that are in everyday society. We see online, for example, a search for Belonging. I see quite a lot in online communities too and the push to just try and find the people that get you online.

Joad Raymond Wren
What’s the definition of utopian or utopia that you’re working with there?

Isabelle Castro
I guess in the sense that if utopia isn’t attainable, they’re not utopian, but maybe that’s also part of the point. We’re in this kind of society now which seems to applaud and reward practicality and making practical steps towards the utopia that they want to see. I see it daily. If someone sees an issue, they actively try to build to solve it.

Share

Like privacy, for example, is something I focus on quite a lot. They see a decline in private, like privacy online and therefore they talk about it and discuss it, sometimes philosophize it, but then also build something, build a startup or try to build something to bring what they want into reality. And I guess that is kind of the real difference between network states in a way and “utopias” because some of them, they do have kind of views that some would see as unattainable. Just the other day I was talking to someone who thought they would never work because of a reliance on national governments.

So I guess in that sense it’s utopian, but in another sense it’s not because there’s a desire to kind of make it so that it can actually happen.

Joad Raymond Wren
You see, in a way I think it’s useful to move the word utopian away from questions of practicality and attainability. And you know, so the sense of the word as it is generally used today it’s something that’s impractical and expressing a set of unrealistic ambitions and therefore in some ways excessively idealistic. Someone who is utopian is a kind of dreamer and that kind of dreaming takes them away from the real world of politics. So, you know, one would call somebody a utopian in order to dismiss their ideas. My ideas are ambitious. Your ideas are utopian. And so on. And I guess one of the things that I’m trying to do in the book is not just to invert that and say, let’s all be more utopian.

I mean, I think it’s fine if people want to be you to want to be that utopian upside down in a sense, but say, okay, why don’t we think more in terms of the longer term tradition of utopian thought in which it’s actually about deconstructing. Oh, I’m not sure I like that word. It’s more about dismantling our mental apparatus and our sense of the values that we have that underpin the society that we live in. And maybe the disparity between those ideals and how they’re realized in society, the way of doing that historically has been to imagine a society that is not just better, it’s not just got, you know, bigger TVs or better healthcare. The health is a really important part of all of these utopias, I think, and it’s easy to lose sight of that.

But actually they’re all founded upon a single principle or set of principles. So the purpose of utopian thinking in that sense, in that longer tradition is not to find a problem in society and say, I know, I’ll imagine that there’s a patch to fix it. There is a problem with the loss of privacy in terms of online engagement. Find a way of enhancing that privacy. They know, they’re saying, let’s imagine a state in which privacy is the core value. How do you build that state from the ground up? What do you do with gender relations? What do you do with marriage? Do you have to abolish marriage or not? Do you have to abolish private property or not? What do you do with houses? Do houses all have to be identical?

So in a sense they’re thinking that their thinking is conceptual rather than practical. But for them, I think the questions of practicality need to be sidelined for the most part in order to, to think that way.

Isabelle Castro
Okay, interesting. It see, yeah, it seems like a mechanism for critical thought, which I mean I see critical thought declining. You might not agree, but in my kind of experience and research, for the most part it is part of that is because of over reliance on AI and increasingly more so.

Where do you see the utopia mechanism as a tool for critical thought having a place today? And I understand that your book leaves off at kind of 2017, 2018. But even today, do you see there is a discourse that allows for this type of critical thinking?

Joad Raymond Wren
Well, I taught at a university for far too many years. I think I’ve probably witnessed the decline of critical thinking, but partly at the hands of a government and a set of managers who wish to introduce a culture that could be assured by measurement rather than openness and open endedness, and partly, I think, probably through the appearance of various technologies, the shortened levels of engagement.

I saw critical thinking in the context of the classroom. And I’m sure there is fantastic critical thinking in all kinds of other institutions that decide to kind of pay for the space for that. You know, occasionally I watch a TED talk and I think, wow, that is absolutely window opening. The problem is, I guess that thought is becoming much more disparate and being drowned out by other kinds of thinking in the context of the classroom.

I taught a module on utopias for about 15 years, I think, and the book isn’t that course, but the book is the consequence of talking to students about utopias. And there was some overlap between the texts in the book. And I think what was really interesting about teaching that module was unlike everything else I taught in that it didn’t have any kind of prescribed trajectory of where it’s going to go. And this is what you need to know. What I did was have a set of ways in which to prompt or provoke students into articulating models that they could use to talk about these kinds of texts. And it was probably the most rewarding teaching I ever did. But I think that’s a consequence of. Partly a consequence of two things, is that the utopia is quite a small genre.

It’s not the same as trying to, you know, teach plays or pastoral. You can probably kind of, in the course of a semester, four months, read enough books for it to kind of make sense from the beginning to the end. So you’ve got some kind of grasp. It may not be massively granular, but you’ve got some kind of grasp of the whole. And if you put students in an unstructured classroom and get them to and kind of monitor and encourage and facilitate that engagement, they come up with ways of managing and describing and analyzing those texts and those historical projects. That’s a space for critical thinking. It happens. It. It happens because.

Because the government pays for it and because it doesn’t have to go anywhere and Again, the assessments were always by far the most interesting of the assessments that I marked because partly because I the, it was a very open ended kind of assessment and partly because I invited them to undertake acts of creative writing to demonstrate a critical engagement with the history of utopian writing and utopian projecting. They weren’t just allowed to kind of write autobiography or anything. It had to be kind of generically informed. And they did some really interesting work. And I think this, one of the ways in which our. Thinking has allowed itself to live within parameters, whether we’re conscious of those kind of parameters or not, is by finding ways to do things that facilitate what we want to do too much. And AI is an example of that.

You know, I’m, I’m probably hopelessly somehow over the past decades I’ve become hopelessly old fashioned and I wasn’t always the case. But I listen to records on vinyl. I make music. None of it. I mean it’s all analog. It’s all about analog. My synthesizers are all analog synthesizers. I can’t come up with a sound and find a way of generating it digitally. I have to use plugs and resistors and so on. But what I value in that, and this is why I think that there’s some kind of harmony between that kind of practice and utopia, is that ultimately I’m fighting against the material world in order to do it. It’s not simply a case of imagining something in a particular way and finding the resources that will make it. It’s this is the stuff I have.

How do I get it to do something interesting? Oh, it’s fighting me. I can’t do that. But actually I quite like the thing it’s making me do. And I guess something similar is maybe happening in how we think about utopias. Again, the attainable is. Maybe thinking in terms of attainability is a bit less interesting than challenging our own creative or critical impulses by engaging them in a particular way and finding out where our limitations are. Maybe that’s the analogy.

Isabelle Castro
Okay, that’s super interesting. And it kind of like I had a question around kind of More’s and Plato’s utopias, which kind of depends on setting out kind of rules. And like as you mentioned, that people just stick to the thing that they’re good at and they can’t really do anything else. Which is really interesting for me because I’m always like writing about people who are kind of. And crypto within crypto there’s a space of people who want abundant freedom and deregulation and all these kind of things. So thinking about that as. And you framing it like that really puts it into context for me in terms of like working out the utopia within the set of rules that is given to us. How do you see? Well, I guess you’ve kind of answered this in a way already. Like the idea how.

The idea of kind of freedom in our current day and age and how that has changed maybe in past decades and centuries where utopia has been thought of. How does that shift? How does that bring kind of usefulness, the use of these kind of parameters in order for us to engage in utopian thinking? I guess that’s how I’m going to frame the question.

Joad Raymond Wren
Well, I guess the model of freedom with which we currently live and which we idolize to some extent is essentially a form of freedom of choosing from an infinite array of choices and leaning too heavily into that as a model of freedom, ends up with a tradition of thinking about utopia which says, I wouldn’t want to live there. Kind of not the point, of course. I think probably within, you know, the history of thinking about freedom as performed in utopias, the emphasis is much more upon a kind of freedom as a form of internal discipline. It has to be said that on the face of it, many utopias are not particularly interested in freedom for them because they’re not. They tend not to think about individuals often don’t have characters.

I mean, the strongest character is often the alien or the outsider who arrives in and observes what happens. And. And if you, in the context of a narrative, if you don’t have characters, then you’re kind of minimizing appetites and will and passions and so on anyway. And maybe that inclines the writer of utopia to think in terms of. And again, there may be analogy with modern society here. Think of humans as a kind of resource, as kind of fleshware for making the perfect society. There are exceptions to that and important exceptions. I think Ursula Guin’s the Dispossessed is a kind of neglected masterpiece of anarchist utopia from the 1970s. William Morris’s news From Nowhere is very much focused upon the nature of freedom.

And it’s the kind of freedom in which self expression and choosing one’s education, choosing the life one wants to live, how one spends one’s time, is precisely constitutive of the kind of world that he imagines that this would create. But for the most part, I think if one were to try and extrapolate a notion A model of freedom, putting Plato to one side from the tradition of writing about utopias, it would be that there is a kind of satisfaction to be obtained through the disciplining of the self, so that extraneous desires and passions don’t overwhelm you. I mean, I think it’s a kind of almost a stoic philosophy, and it’s probably quite useful one to. To have today if. I mean, if you go shopping in an actual physical supermarket rather than online and being able to discipline one’s passions.

And to say, what do I actually know about this olive oil? When presented with a choice of 1500 olive oils, what is the right. Not which is the right one, which one do I want, but what is the right kind of choice that I should be making at this point? I mean, that’s a kind of recognizable philosophy, I think. But that’s the kind of understanding of. Of freedom, I think that is probably embedded in. In many earlier utopias.

Isabelle Castro
I just want to go back a bit to what you said about how it doesn’t really. A lot of these utopias, the literary ones, don’t think about the individual. They think about the community and the society as a whole. Why do you think that is? Yeah, I’m just gonna leave it because I feel like that is enough of a question to handle, I guess, because.

Joad Raymond Wren
They’re thinking about community as a whole because the emphasis is not upon individual satisfaction, but collective satisfaction. That. And I. I don’t think they are for the most part. Some of us, Some of them are, but for the most part, they’re not utilitarians. They’re not thinking about the happiness, the greatest happiness of the greatest number. They are removing individuality in order to think about collective structures. And that can make them unsatisfying as works of fiction. I think you can make them rather flat as works of fiction. But in terms of thinking about questions like health and happiness. Happiness is a weird example to give, but they are. I mean, I think some of them are precisely focused upon happiness.

But by looking at the kind of happiness that is rooted in equality and, you know, ultimately homogeneity, looking at the kind of happiness that might be available in a society in which everyone is equal. And maybe there aren’t a great number of opportunities for differentiation. Everyone is living in the same house. There’s no. In the same model of house, there’s no scope for envy. There may be freedom of choice in relation to marriage choices and utopias were big on. On marriage choice and in. On gender equality before Most of European society was they. I guess the question of. Of individuation, it would simply be a distraction. And if you contrast that with dystopias. I don’t want to talk about dystopias, but why am I mentioning dystopias?

But dystopias are always about an individual who’s unhappy, whereas utopias are a completely different kind of book, a completely different kind of fiction, a completely separate tradition. And they’re kind of. They’re not about you. They’re not about you and me. They are about abstract constructions that we might be able to fit into.

Isabelle Castro
Mm. Okay. I guess this comes back to the kind of practicality of it because, of course, like, I mean, again, this is where we come at it from very different angles in our research. In terms of, like, a lot of the things that I cover in the wider tech roots of network states are very individual focused and really about the individual. And the idea for a lot of people of this collective thing where you don’t have maybe private property or something like that is like, exactly what they’re rallying against. So it’s this really kind of interesting way to see how that. Because, I mean, I know you say that it is kind of like a. It’s imagined. An imagined scenario, but I would say that some of the things that have been imagined in these scenarios, in these utopian scenarios have come to play in real life.

Joad Raymond Wren
Oh, absolutely. So, I mean, you’ve got Etienne Cabet writing in 1840, describing this fictional society called Icari. The furniture has rounded edges, so children don’t hurt themselves on it. There’s. The traffic is controlled by policemen with trumpets. There is gas street lighting. The hospitals have. He doesn’t describe how this happens. Have music invisibly played in all of the rooms. 1840. Skip forward to 1888. Edward Bellamy is Looking Backwards, which is massively popular book. Very, very influential. Again, people have music in their houses that they’re able to listen to through something a bit like a telephone, which is all been played live elsewhere. So there are four recording studios around the world. So it happens 24 hours a day. And they’re able to dial into this and listen to it through something like a telephone. They’re able to. The volume so that they can.

They can, you know, lie in bed and listen to it without disturbing other people. And. And the. The variable resistor, the rheostat has only recently been invented by 1888. So it, you know, what else does he have? A retractable covers for the. The sidewalks so people don’t have to carry umbrellas around with them. Electrical, street lighting, debit cards. He uses the word credit card before the thing exists at that point. So there are extraordinary inventions riveted in some of these books that then are realized. My favorite example is the neighbor Peloni’s Looking Backwards. He describes a department store which has floating walkways and it’s made of steel and brick, lots of glass. And Wyman, George Wyman, read this book and designed shortly thereafter, a few years later, the Bradbury Building in Los Angeles. Again, similar of construction.

So, you know, long before the international style of modernism at all. Open steel, glass and brick construction, which is then used in Blade Runner as the. For the apartment in which J.F. Sebastian, the genetic designer, lives with all of his robot toys. And also where Rutger Haa does his final scene on the roof as. As he dies. So the kind of. The Most influential Late 19th century utopian novel spawns the most influential dystopian film of. Of all time. So there are. There are. They. They do affect things kind of inadvertently or inadvertently, practically or impractically. It’s a different. It’s. Is it. It’s not a different kind of. It is an inadvertent product of thinking in a profoundly speculative way.

Isabelle Castro
Okay. Yeah. Because, I mean, I think there’s a. There’s a line. I was just trying to find it in my notes. “Growing up in utopia is a good way to make you an incrementalist,” from Akash Kapur’s memoir on Auroville. I thought that was very interesting and kind of, I guess I want to ask how does that relate to the lack of attainability that you see in kind of the more literary utopias?

Joad Raymond Wren
Yeah, I wonder whether on reading that line, one needs to revert back to the question of whether to engage in utopian thinking necessarily needs to result in a schema for change. It might result in a desire for change. It might result in a set of ideals for change. It doesn’t necessarily furnish you with the answer to the question of whether you should do it in large structures or by increments. Paradigmatically, I think it changes thought, but not necessarily political practice. And there are many ways, I think, in which notwithstanding street lighting and public toilets, and so on. there is a sense in which to pick bits out of utopia and apply them to your everyday life is to call disaster without thinking about the structures in the broader structures in which those details live is to invite disaster.

One of the few people to do that successfully, I, I think, was Vasco DiCoroga, who’s a contemporary of Sir Thomas More’s, who reads it and absolutely astonishingly decides it would be a good idea to try and implement this in Mexico. I mean, he leaves out all of the, you know, the stuff that would have been deemed weird, such as naked courtship, the idea that you might actually see the person you’re about to marry naked before you marry them, or the ordination of women. Absolutely inconceivable in the 1530s. He leaves aside all that stuff.

But of communal living, he tries to import one of the strangest books of the early 16th century into. Into real life in Mexico in the 1530s. Notwithstanding that often picking and choosing elements of utopias is. Is to. And to try and implement them is. Is to invite ridicule. Because what the books invite us to do is to think a totalizing level, whether that’s going to be practical or not. I mean, I have a question for you in terms of. Because I’m still thinking about that question about individuals and you’re describing the culture you work in as providing for a kind of individualism.

Isabelle Castro
Yeah.

Joad Raymond Wren
Do you think that. Do you think that individualism has grown, let’s say, or the expectations of individualism have grown over the past few decades and is. Does that reflect increasing as being increasingly individual? Is it possible there’s a total disconnect between the extent to which we’re individuals and our sense of ourselves as individuals and having individual needs that we need supplying? Or could we. Is it possible there’s actually more homogeneity than we recognize?

Isabelle Castro
I feel like there is a. This is kind of the question that a lot of these communities and even kind of governance projects are working towards with their experiments, is how do you preserve individual rights and the desires of the individual and the autonomy of the individual. I do think there is this kind of line, a tightrope walked between how do you preserve all of these things that a lot of these people believe in and yet build a community that is functional and can grow with a community mindset and incentivize engagement in the community.

Because whether we like it or not, even if we want to be individuals, even if we want individual autonomy we are part of society which is more of a collective mentality. So it’s something that we have to engage with and think about.

Maybe this is why there is a kind of leaning towards the founder-led model of these societies because it’s something that people can kind of understand within the context of a startup where you have to have autonomy and it can be good to have your own ideas but youre thinking collectively in the sense that youre values are aligned with a common purpose. Because, it must be said, a lot of the people that I’ve kind of engaged with personally are coming from Silicon Valley and the whole mentality that comes with that, which I did actually want to go into a little bit because I mean, for me Silicon Valley is the most active utopian builder of our era. You’ve got network states led by people who kind of cut their teeth out of there, of course, you’ve got seasteading coming out with that. I don’t know whether you’ve looked into that and like projects for building rockets to go to Mars and kind of Mars communes and all this kind of stuff.

At a certain point these ideas were seen as utopian. These ideas were seen as unattainable. These are kind of like moonshot ideas, but they do come into reality. And I don’t know, I feel like this relates to the whole kind of incrementalist and like what active role does utopian thinking have in society and our progression of society?

Joad Raymond Wren
You see, I can hear echoes of old utopias, utopian science fiction films, yeah, 60s in particular in this. And I can see how this might fit into the, the tradition of the utopian genre which is always looking for other worlds. So you know, back in Plato’s day it’s a mental construct in More’s day it’s an island that you discover because there’s still plenty of the world being discovered, rediscovered at this point, eventually it has to shift to alternative time zones or alternative universes. If we’re going to find an undiscovered land and visit it and find out its internal logic, then we need to find ways of it being. An existence. So it goes from being. It’s out there to it’s. It’s. It’s in the future or in a different reality.

I can see how that feeds in to the sense that planning a moonshot and a moon station, planning a kind of virtual community in an online world that has some sign of some kind of simulacrum of real existence, if you’re not. Even though you’re not bodily present there, I can absolutely see how that fits in to an historical tradition. I do wonder whether the impulses behind that, whether the motivation and whether the conceptualization is all that sympathetic towards the likes of, you know, Plato and more and Margaret Cavendish and Sarah Scott, or whether it isn’t, in fact, just a very ambitious mode of contemporary thinking made possible by improvements in technology and increasing wealth inequality.

And in some ways, it seems to be pulling away from the sense of collective responsibility and collective engagement and the sense of responsibility for universal health care and gender equality that inhabits earlier traditions of utopian thinking. Again, it seems to me that’s. It’s only utopian thinking insofar as we think, oh, it is something that’s so ambitious, it’s on the edge of the possible.

It was hot in London, that kind of heat that nudges Londoners out of their houses, spilling out onto the streets to enjoy their brief slice of Euro summer.

Within this backdrop of sun and long evenings, I’d found my way to a party in a converted warehouse in the shadow of the Shard, near London Bridge. A smattering of founders and developers dancing to ABBA’s Gimmie Gimmie Gimmie a man after midnight, providing the soundtrack to my rant on the decline of privacy in the UK.

It had been a week since (now ex) PM Keir Starmer had announced the Labour Party’s intention to extend the Online Safety Act, mandating age verification from spring 2027 to enforce a social media ban for under-16s.

On the surface restricting access to social media for children sounded like a great idea - even I could feel the effects of too much scrolling - but its implications meant a tightening on internet freedom for all citizens of the UK. Once the ban is in place, in order to access social media platforms, every social media user would have to upload identifying information, like passports, to access their accounts.

In 2025 a similar restriction had been implemented for websites with sensitive content, which I went into in a previous newsletter last year. The increasing sophistication of AI used to assist in harnessing already vast amounts of online data was already making those that I spoke to about both AI and privacy tech nervous.

After a day with friends who had already told me, “Take my data! I don’t care” I thought maybe I was overreacting. But, in the warehouse, I had met a consultant who was working with some of the biggest companies in the world with their AI integration. He too, was concerned about the amount of data already online, and how the restrictions might add to it.

But I’ll go into all that and the nuance of the ban in more detail below.

The reasoning behind it

The social media ban is the latest application of a sweeping set of laws titled the Online Safety Act that allows the regulation of online content, making internet service providers responsible for the content shared on their platform. When applied to social media, it means that social media platforms will be responsible for checking the identity documents of their users, in order to allow access.

The ban itself is nuanced because of its positioning as a way to target huge problems with children’s usage of social media and the internet. Honestly, if I had children, I too would be very concerned about what they are exposed to online.

Starmer said in his announcement of the social media ban that it comes as a way to address the declining mental health in children, caused, in part, by heightened social media usage. Many of the reports cited, including a consultation carried out by the government itself in 2026, are quick to highlight that defining causation has been challenging. A report carried out by the House of Commons Science and Technology committee in 2019 stated that it was difficult to determine whether mental health conditions drew children to spending more time online or whether the time itself exacerbated problems. However, a report submitted by Bernardo’s Children’s charity in 2018 found that mental health disorders in children rose from 12% to 27% when using social media for more than 3 hours a day.

Share

A 2026 report by Ofcom cited in the consultation document found that “the impact of social media is nuanced” and “negative impacts are small when averaged across the population” it indicated the features of the platforms made it “harder for children to consume social media in moderation.”

Then there are the problems of bullying and exposure to harmful content. The House of Commons 2019 report noted that social media could amplify existing problems of bullying and grooming while Ofcom’s media use attitudes report of 2026, found that of the 23% of children that reported “nasty or hurtful” behaviour happening to them personally in the past year, it was split between 47% online and 43% face-to-face.

Therefore the ban was deemed necessary by the UK government because the “risk outweighs the benefits.”

But although I’m of course on board with addressing all of these issues, as are all the people I have talked to about the ban, their concern focuses on the implications of such a quick implementation of a blanket ban.

“A ban on under-16s is impossible to implement in a way that only affects under-16s,” said Joe Andrews, CEO of Aztec Labs, a company that develops blockchain backed privacy-centred products. “The only way to keep a teenager off a platform is to check the age of everyone signing up, so a rule aimed at children alone becomes an ID requirement for the whole country.”

Data honeypots

In 2025, the invocation of the Online Safety Act led to implementation of age restrictions on websites with adult content.

This first application brought with it the rumblings of concern from privacy advocates. Citing heightened instances of online hacks of personal data, jumping 79% in the last five years according to the latest Identity Theft Resource Centre report, they reasoned that the uploading of sensitive identifiable data like passports could only do more harm than good.

“Every platform will now require you to hand a passport or a face scan to a verification vendor, and every one of those vendors becomes a honeypot of real IDs waiting to leak,” said Joe.

“We saw it with the Online Safety Act checks last year, we saw it again when a provider connected to Discord leaked tens of thousands of government ID images, and we’ve seen it countless times over the years. Millions of adults take on the real risk, identity theft and fraud, for a measure that determined teenagers could easily route around with a VPN anyway.”

Raido Saar, founder and CEO of digital ID provider, ComplyOnce noted that already criminals had accessed the data of 200 million Pornhub users in December 2025, compiled of email addresses, viewing activity and location.

“This data is not only private, but this is really delicate,” he said. “This data is valuable on the dark web, and is really usable for extortion and quieting your political opponents.”

Historical abuse

The implementation of an age-verified gateway to social media platform brings with it an implication that stretches beyond the potential for extortion that Saar noted could follow data collected by adult websites. Privacy advocates pointed to a number of historical instances where online data collection has already been harnessed for identification, surveillance and influence using legitimate data collection techniques.

The first, highlighted by Aztec’s Joe, was in the release of AOL search logs in 2006. The company released a large excerpt of search queries without related names and email addressed, intended for independent research. However, numbers associated with the accounts related to the searches allowed reporters from The New York times to track down and identify associated users. AOL took down the data, by which time it had been copied throughout the internet.

Search queries were described as “highly personal” and embarrassing in the class action lawsuit following the release. Among the information surfaced were searches related to health concerns, escaping abusive situations, sexual assault and inflicting harm. The company’s release of the data was found in breach of electronic privacy laws and settled in 2013.

The release of the data highlighted the issue of mass data collection online. The risk, he said, is that individuals have no control over the data collected yet carry the cost of it being released to the public. “Your location history gets sold and shows where you live and who you visit,” he said. “The shopping you did gets fed into a pricing model that sets your insurance. A dataset leaks once and follows you for a decade. You can’t reissue your face the way you reissue a card. None of this stems from you having done anything wrong.”

Share

“You can’t reliably anonymise behaviour, and once it’s pooled somewhere it’s one breach from a name. The safest data is the data you never collect.”

The second, cited by Roxana Nasoi of Logos, is in the Facebook Cambridge Analytica scandal of the 2010s. The research firm, Cambridge Analytica, was accused of using personal data connected to 87 million Facebook profiles to psychologically target users and assist the presidential campaigns of Ted Cruz and Donald Trump in 2016. The company was also accused of using the data to assist the influence of Leave.EU’s Brexit campaign although later investigations found no evidence of misuse.

Restricted freedoms

All of this data collection is one thing, but when put into the context of how global internet freedoms are headed, it really makes you think twice about what you want to share online.

Of the 5.5 billion people who have access to the internet, Freedom House’s latest Freedom of the Internet report reads, 81% live in countries where people have been imprisoned due to an online post. Although the US, Canada, the UK and much of Western Europe have kept their status of digitally “free”, increased restrictions imposed by governing administrations have brought their rating close to losing that classification.

The implementation of AI tools has increased governments’ capacities for surveillance. According to the Carnegie Endowment’s AI surveillance index, both America and China have deployed AI surveillance technology around the world. “Liberal democracies are major users of AI surveillance. The index shows that 51% of advanced democracies deploy AI surveillance systems. In contrast, 37% of closed autocratic states, 41% of electoral autocratic/competitive autocratic states, and 41% of electoral democracies/illiberal democracies deploy AI surveillance technology,” the paper stated. “The most important factor determining whether governments will deploy this technology for repressive purposes is the quality of their governance.”

The application of surveillance measures is often dressed up as positive enforcement as a way of increasing methods to catch online criminals. In a recent hearing regarding the Canadian C-22 Lawful Access Bill, a bill that would allow the Canadian government to enforce back doors into digital service providers’ encryption, Canadian Conservative MP, Dane Lloyd, who opposes the bill, stated that “Law enforcement and national security experts have long argued that they do not have the tools to effectively go after terrorists, organized criminals and child predators, who are coordinating a lot of their activity online.”

Similar arguments were made in the implementation of the 2016 Investigatory Powers Act in the United Kingdom, which, it resurfaced last year, allowed the UK government to secretly enforce similar surveillance mechanisms.

This justification is enough for some, who opt instead for “centralization and convenience.” In a world where internet freedom is upheld, what does it matter if the government knows about what someone does online, as long as they aren’t a criminal? However, in the world we live in, where online privacy is in continuous decline, the presumption of “innocence until proven guilty” turns into a presumption of guilt and collection of data to continually prove innocence.

Designing for consent

So now I’ve gone through the pains of outlining all that, and you, taken the time to read it, what then is the solution?

“The real battle should be around protecting consent,” said Roxana. “Privacy is consent to access what makes you who you are [...] the erosion of consent [to internet data] is the erosion of online privacy, which has triggered a decline in internet freedoms over the past 20 years, and has impacted civil society.”

There are technological solutions being built to protect consent to data, which you can read about in more detail in my coverage of this in Sandmark. But, in the words of Doro Unger-Lee, Director of development at Corpus Christi College Cambridge, who organised a roundtable to discuss privacy-preserving technology at the House of Lords in November 2025. “The gap is governance, incentives, and political will.”

Roxana told me about the round table, that there were experts on data ownership, academics, technologists and biometric identity developers, all there to discuss a privacy preserving Digital ID solution.

The implementation of the Online Safety Act restrictions came two months before the UK government announced the roll out of a digital ID scheme that would be made mandatory to prove a right to work. Introduced without a public consultation, the announcement was met with criticism. An eight week consultation period was delayed until March 2026.

Given the government’s latest moves to ban social media without age verification, “some sort of digital ID is kind of inevitable,” she said, as it would make it easier for internet users to verify their age online. While the UK has stated that digital IDs “can offer privacy protection,” little has been released around how they will design the IDs to ensure privacy.

Share

Although the round table to discuss privacy preserving solutions was held in the House of Lords, “None of the Lords attended,” said Roxana Nasoi, spokesperson at Logos, a decentralised community that develops blockchain based privacy technology. “The Baroness [Manzila Uddin] who was chairing was completely overwhelmed because it was highly technical.”

In a research briefing published in March, the government highlighted that the ID would be built “in house” leaving little room for iteration, however, it seems like there’s little interest or understanding in making sure privacy protection will be a feature.

“It’s going to open up so many issues. Maybe some of them we haven’t even grasped yet,” Roxana told me.

And yes, this is about the UK, but as Roxana told me later, things passed by the UK are likely to have knock on effects on how the same problems are treated in other jurisdictions.

Directions of travel

Walking home from the warehouse that hot night, ABBA still ringing in my ears, I still thought - am I overreacting? The friends who told me they were happy to wave away their data have a point, there is a real ease in letting the platforms and the government handle the admin of proving who you are in order to access what you want online. For most people, most of the time, for now, nothing bad comes of it at all.

But it’s about the direction of travel that has me worried. Everyone agrees that the stated goal of the current ban is a good one, something that should be addressed. However, the children it claims to shield are the ones who will inherit a country where access to information online is dependent on who you are.

A House of Lords round table filled with experts to discuss how you protect individual rights while you do it, and not one of the people meant to build and govern the system itself bothering to turn up, tells you roughly how much of the governments’ thought is going into what this means further down the line.

FAQs

Q. Why is the UK banning social media for under-16s?
The government frames it as a response to declining child mental health and exposure to bullying and harmful content, citing evidence such as a Barnardo’s submission that found mental health disorders in children rose from 12% to 27% among those using social media more than three hours a day. Officials concluded the risk outweighed the benefits, though the cited research repeatedly notes that causation is hard to establish.

Q. Does the UK social media ban require everyone to verify their age?
Yes, in practice. Because a platform cannot tell who is under 16 without checking, enforcing the rule means age-verifying every user, which turns a measure aimed at children into an ID requirement for the whole adult population, a point made by Aztec Labs CEO Joe Andrews.

Q. What are the privacy risks of online age verification?
Each verification vendor becomes a concentrated store of real identity documents that can leak, and online data breaches have risen sharply, up 79% over five years according to the Identity Theft Resource Centre. Recent cases include around 200 million Pornhub users exposed in December 2025 and a Discord-linked provider leaking tens of thousands of government ID images, data that is valuable for fraud and extortion.

Q. Hasn’t collected personal data been misused before?
Yes, and the essay points to two cases. AOL’s 2006 release of supposedly anonymised search logs allowed reporters to re-identify users from the queries alone, and the Cambridge Analytica scandal saw personal data tied to 87 million Facebook profiles used for political targeting, illustrating that data pooled for one purpose tends to escape it.

Q. Is the UK introducing mandatory digital ID, and will it protect privacy?
The age-verification push arrived alongside a digital ID scheme proposed without public consultation, and advocates argue some form of digital ID now looks close to inevitable as the route to verifying age online. The government has said digital IDs can offer privacy protection but has published little on how, and a House of Lords roundtable on privacy-preserving designs was attended by the technologists but none of the Lords themselves.

But the other reason was due to the footnote at the bottom of the product launch document:

1. Mythos-class models are a tier of Claude models that sit above our Opus class in capability. The first, Claude Mythos Preview, was released in April through Project Glasswing. That is followed today (Jun 9) by Claude Fable 5 and Claude Mythos 5.

2. Fable is from the Latin fabula, “that which is told,” akin to the Greek mythos. The safeguards are what distinguish the two models (Fable and Mythos) and are why we’ve given them different names.

It was quite a candid admission to bury in a product announcement which is why it caught my eye. It basically indicated that Fable and Mythos are not two models, they are one model with two names with the only difference between them being the safeguard.

Fable was the version released to everyone. Mythos was the version released to a small group of cyberdefenders and infrastructure providers through Project Glasswing, a collaboration with the US government, with the cybersecurity safeguards lifted.

Share

On the one hand, it was perhaps a good thing they were almost the same, everyone using the Pro and Max versions of Claude would have access to the same model as Mythos. The problem was the team at Anthropic had spent a good deal of time writing and talking about how powerful and potentially dangerous Mythos could be before they decided to release it.

Not that kind of danger, silly

It is worth reading what Anthropic actually said about Fable 5 because the launch post spends a great deal of its length arguing for its own product’s danger. Fable 5’s capabilities, it says, exceed those of any model the company has ever made generally available. Without safeguards, those capabilities could be misused to cause serious damage, particularly in cybersecurity, where Mythos class models can carry out what the company calls agentic hacking, meaning the full chain of an attack rather than one step of it.

This was backed by the comments of another crypto cybersecurity I had reached out to for my own reporting on the crypto vulnerability side before the launch. The threat wasn’t in the capabilities to discover vulnerabilities, they had told me, it was the ability it created for individuals to weaponise the vulnerability quickly.

The gap between a vulnerability existing and a vulnerability being exploited used to be filled by weeks of work. What Mythos 5 (and therefore Fable 5) does, on Anthropic’s account, is shrink that gap to whoever happens to be signed in on a Tuesday afternoon. Fewer than 5% of Fable sessions get redirected to Opus 4.8 for safety reasons, which means more than 95% of the time, the person typing into Fable is, for practical purposes, typing into Mythos.

Subscribe now

Anthropic said it ran an external bug bounty that produced no universal jailbreak in over 1,000 hours of testing. However the UK’s AI Safety Institute did make progress towards one within a brief initial testing window. So, Anthropic concluded it is likely impossible to completely prevent universal jailbreaks. This is, by any reasonable standard, a company telling the public that this is a weapon adjacent technology, and that it knows it.

Three days after the launch, at 5.21pm ET on June 12, the US government sent Anthropic a letter. Citing national security authorities, it ordered the company to suspend all access to both Fable 5 and Mythos 5 for any foreign national anywhere, including Anthropic’s own foreign national employees. Because there was no practical way to comply with that, Anthropic disabled both models for everyone. Access to every other Claude model stayed the same.

Anthropic’s response was to argue that the danger was not that kind of danger and simply a “misunderstanding”.

They explained that the jailbreak the government had apparently been shown, essentially consisted of asking the model to read a specific codebase and fix any software flaws, a capability widely available from other models, including OpenAI’s GPT 5.5 used every day by the people who keep systems safe.

Recalling a commercial model over a finding like that, Anthropic argued, would essentially halt all new model deployments for every frontier model provider.

Both of these positions were taken by the same company about the same product in the same week. The first established that the model is dangerous enough to need a defence in depth architecture, a 30-day data retention policy to research and mitigate jailbreaks as well as a tiered access system run in partnership with the US government. The second argued that a specific instance of the thing that architecture exists to catch does not really count.

To me, it speaks to a lot of how the story around AI is panning out online and its general development. We hear, it seems weekly, about the existential threats that powerful AI poses to the world, by the same people who build it.

In the launch document, yes, Anthropic was transparent about the threats the model posed - that is good. But while stating that it was almost impossible to stop these threats happening the company launched the model anyway.

Funny weather we are having

Roland Barthes wrote about how myths work in ordinary, present day life as a way of talking about things that makes them seem natural rather than chosen. A myth, in his sense, takes something that exists because of a series of decisions made by particular people and talks about it instead as though it had always been this way and could not be otherwise. The history behind the thing disappears, and what is left becomes almost a fact of nature.

We are now well versed in the idea of the singularity, and the idea that the rate of technological advancement will reach an irreversible point of no return. Honestly, I believe that AI is coming and we, the average person, can’t really stop it.

At this point, however, we are being fed models that are still being built and deployed by companies managed by humans caught only within a fierce competition rather than directed by the models themselves. Using the Barthes framing, the use of the “AI is coming, you can’t stop it” narrative by the people who make the models takes a sequence of choices that they made and transforms it into an unstoppable process, something closer to the weather than a roadmap.

Share

The use of that narrative as justification to launch products, in this instance, and possibly many more, essentially acts as an excuse for companies to wash their hands of the responsibility that comes with that descision.

Decisions Decisions

What is going to be interesting is how this all plays out, and how long it will take. Given some of the moves in the past year, its highly likely Anthropic will be able to sufficiently explain the “misunderstanding”, reach some kind of compromise with the government and relaunch some version of Fable in the next week or two (although I wouldn’t be surprised if it’s basically an Opus under another name).

But at this point of pure speculation, I find some parallels with how cryptography and encryption was treated back in the 1990s.

In the early 1990s, the government classified strong encryption software as a munition, the same export control category now being used against Fable and Mythos. Exporting code containing certain kinds of encryption required the same kind of licence required for tanks and missile parts.

Phil Zimmermann was investigated for distributing PGP, a strong encryption program that provides cryptographic privacy and authentication for data communication. Daniel Bernstein, a Berkeley mathematician, sued the government, with the Electronic Frontier Foundation, for the right to publish an encryption algorithm and its source code without asking permission first, arguing that it was a form of freedom of speech.

It took until 1996 for a court to rule that code counts as speech, and an executive order that same year started moving most encryption off the munitions list. Full removal of export restrictions on retail encryption did not happen until 1999, several years after the controls were first seriously contested.

None of this stopped encryption being developed though. It got built in the background on open source networks using international development, so it could bypass the restrictions put in place by the US. PGP escaped its control problem because it stopped being one company’s product to control. Once it was out in the open other people kept building on it and by the time the export rules caught up there was nothing left to export, the code was already everywhere.

If frontier AI development goes anywhere similar, a directive aimed at one company’s commercial model might end up pushing the next wave of capability toward open, distributed projects. Decentralized AI has mostly been a small sector so far, built by people who already think the industry is too concentrated. A government that can periodically switch off the market leader’s product is the kind of thing that makes the argument for decentralisation argument look a lot more practical.

This doesn’t make the original problem around whether Fable was too dangerous go away though. If development runs towards a more decentralised framework, who then will decide when it's become too dangerous or not?

_FAQs

_{Q: Why did the US government suspend access to Claude Fable 5 and Mythos 5?
A: On June 12, 2026, the US government sent Anthropic a letter citing national security authorities and ordered it to suspend access to both Fable 5 and Mythos 5 for any foreign national. Because Anthropic had no practical way to comply with that narrowly, it disabled both models for everyone, while access to other Claude models stayed the same.}

_{Q: What is the difference between Claude Fable 5 and Claude Mythos 5?
A: By Anthropic’s own admission in a launch footnote, Fable and Mythos are essentially one model with two names, distinguished only by which safeguards are applied. Fable went to the public, Mythos went to a small group of cyberdefenders and infrastructure providers through Project Glasswing, with cybersecurity safeguards lifted.}

_{Q: How did Anthropic respond to the government’s suspension of Fable 5?
A: Anthropic called the suspension a “misunderstanding,” saying the jailbreak the government had been shown amounted to asking the model to read a codebase and fix software flaws, a capability already widely available in models like GPT-5.5. It argued that recalling a commercial model over a finding like that would effectively halt new model deployments across every frontier AI provider.}

_{Q: What percentage of Claude Fable 5 sessions get redirected for safety reasons?
A: Anthropic says fewer than 5% of Fable 5 sessions are redirected to Claude Opus 4.8 for safety reasons, meaning more than 95% of the time, a user interacting with Fable is, for practical purposes, interacting with the more powerful Mythos-class model.}

_{Q: Could AI export controls push development toward decentralized AI?
A: Possibly. PGP escaped 1990s export controls once it stopped being one company’s product and became an open, distributed project that other developers kept building on. A government directive aimed at one company’s commercial AI model could similarly push the next wave of frontier capability toward open, decentralized projects that are harder for any single government to switch off.}

Things will be cooling off now so expect to receive newsletters in your inbox weekly as before. Thank you for sticking around!

The below came up within this short hiatus, and while it is focused on network states I think the general theme speaks to our increasingly online lives. If any of it resonates leave a comment with your thoughts below.

In one of my network state group chats on Telegram, a debate has been raging. It’s been going on for a few months now really. The conversation centres on what defines a network state.

On one side, there are people who are content with virtual governance and economies, saying that this is enough for characterisation. On the other, are those that think territory has a big part to play.

The official definition as set out in The Network State is such - “A network state is a highly aligned online community with a capacity for collective action that crowdfunds territory around the world and eventually gains diplomatic recognition from pre-existing states.”

Later on the book clarifies certain requisites. It needs a crypto economy, a network union rather than a simple community, and, an important part, territory.

“A key concept is to go cloud first, land last, but not land never, by starting with an online community and then materializing it into the physical world.”- Balaji Srinivasan, The Network State.

I bring this up today, because I was back in Network School during this month (which you might remember I visited last year). While I was there this time an event posited the question “Is Ethereum is a network state?”

Put more broadly, in my mind it asks the question “To what extent does digital infrastructure stay online, before it spills into our physical surroundings.”

“A network state is..”

In my time researching network states, the concept of what a network state actually is has been discussed in a number of ways. I’ve been in conversations that asked “Is a multinational corporation (like McDonalds) a network state?”, “Is the Church a network state?” - the answer to both has been “no” for the most obvious reason that they don’t have a blockchain economy (although they are fun thought experiments regardless which make the concept more tangible so I might go into these questions at a later date). Ethereum, on the other hand, is a blockchain and a currency so that classification is a bit more blurry.

Ethereum has been the locus point for a lot of network state ideas. Zuzalu, for example, was the first pop-up city. In its wake, a rash of “Zu” communities have been born, along with the abundance of network states. These all have territories, whether temporarily or more permanent, and many use Ethereum or a derivative blockchain to run their operations.

But Ethereum itself, despite being the umbrella under which they all align, and despite having its own community which hosts a number of meetups globally, is something else. I’ve been trying to work out why.

The parallel polis

The reason why its not a cut and dry answer goes back to the idea of the parallel polis which was first brought up by Czech political thinker, Václav Benda in 1978.

At the time, what we now know as the Czech Republic was still Czechoslovakia and under a totalitarian communist regime. Benda observed the emergence of a new social structure within intellectual and artistic circles which acted as a form of social organisation to “escape” the regime.

Within it, he observed seven “pillars”:

• A group of people that actively advocate for and protect their rights or values.

• A culture that is independent and consists of creation without the permission or support of public authorities.

• An education system and development of scientific research with freedom from the presiding political regime.

• A parallel information system as an expression of the right to the free dissemination of information without censorship.

• A parallel economy, which, at that time relied on bartering and direct reciprocity with trusted individuals.

• The creation of alternative governance structures.

• A parallel version of foreign policy which could legitimise the movement and secure moral, financial and mental support.

The parallel polis that Benda was a part of, Chapter 77, used these pillars from 1976-1992. The group was criticised and repressed with many of its members dismissed from work, forced into exile or imprisoned. Under the communist government, their influence was limited and its members mainly resided in Prague, with the majority of citizens only aware of its existence due to the government’s campaign against it.

However, during the Eastern Bloc Revolutions of 1989, particularly the nonviolent “Velvet Revolution” within Czechoslovakia, some of Chapter 77’s members got involved and helped with the transition into a democracy. Shortly after this common goal was obtained, the Chapter 77 initiative dissolved.

The ideas behind the parallel polis did not.

Share

The modern evolution of the parallel society

Reading the pillars of the 1978 Parallel Polis is interesting in the context of the world we now live in with the internet underlying most aspects of our lives. Web3 and crypto offer the permissionless tools to allow for many of the pillars to be created, while the migration of much of our social lives to social media creates a tool for global connection and information sharing.

Of course, I’m not the first to notice this. Some of the projects since are explicitly parallel polis inspired. In Prague in 2014 a new hub of cypherpunks, Paraleni Polis, was created, with Bitcoin forming a significant part of how they transacted. Subsequent projects, such as the Paralelna Polis in Brataslava, continue to exist today evolving with the development of Web3 tools. Both of these have physical elements to them, through events or cafes - places where people can meet in person.

Where Ethereum comes into all of this is its positioning as a “network of networks” acting as a virtual substrate for others to build on to create the pillars.

Logos, somewhat related to Ethereum, is another project that is building out a set of approaches that underpin the freedoms of governance, information sharing, education and creation. The project’s manifesto is pretty broad - “forkable” - focused mainly on building the tools others can take acting as a platform for others to create their own “self sovereign virtual territories”.

And maybe a virtual territory is enough. As I went into in “Beautiful and Possible” all land is accounted for by existing nation states, perhaps the virtual realm is the only frontier that actually allows for true self-sovereign freedom (unless we count space). As we have seen with the development of online communties, the idea of meeting and organising online based on values without meeting the other people in person, is not such a foreign concept anymore.

But we are physical beings. Unless we are destined for a life of physical solitude but online social abundance (which perhaps some people, and companies, would love) at some point territory, whether permanent or temporary, large or as small as a single building, will come into play if it’s going to have real life impact.

Territory and tangibility

The parallel polis is relevant here beyond the historical context. What Benda was describing existed in physical space. Chapter 77 met in apartments, circulated samizdat by hand, and for the most part at the beginning was confined to Prague. Its members didn’t debate whether to meet in person, they just did. The parallel polis was virtual in the sense it was a parallel idea with alternative structures its members believed in and used but it occupied real territory and that face to face nature was part of what made it function.

This is where the broader question of what extent digital infrastructure can stay purely online before it has to spill into physical surroundings comes in.

Last year I spoke to Lane Rettig, a longtime contributor to the Ethereum ecosystem, about DAOs, the governance structures that are really the embodiment of many of the parallel polis’ pillars. They had, in his view, largely failed.

“They’ve been disastrous, yes. In general,” he said.

Much of this outcome, in his eyes, came down to lack of phyiscality. DAOs had tried to do everything online.

“If you're serious about governance, you need to get the stakeholders physically in a room together,” he said. “It is the only way that humans build trust, period. No exception. There's no other way to do it.”

“Put butts in seats, get people in a room together, have a beer together, build trust together. I think that alone would solve 80% of DAO problems. People who are very compatible online, they just sit down face to face and 95% of issues just vanish.”

Digital infrastructure, even with all the coordination tools Web3 provides, appears to hit a wall when it tries to replace the physical entirely. The DAOs that built on Ethereum had the economy, the aligned community, the permissionless tools and the capacity to build more. What they didn’t have was territory. And that absence, in practice, is what held them back.

Going beyond the conversation with Lane about DAOs, Elinor Ostrom spent decades studying how communities govern shared resources and won the Nobel Prize in Economics in 2009 for what she found. Her core argument was that face-to-face communication is foundational to building the social capital that makes governance actually work. People “who directly interact with one another and with the physical world,” she found, can adapt and modify rules in ways that remote governance can’t replicate.

Nassim Taleb makes a related point in Skin in the Game. His argument is that accountability decreases the further you are from the consequences of your decisions, both geographically and organisationally. Physical stakes force accountability and without them, the incentive to take things seriously just isn’t there in the same way. DAOs, anonymous and remote by design, remove exactly that kind of stake.

Share

The evidence also shows up clearly in online education, which I explored a little in “Classrooms in the age of AI”. Even though the people I spoke to were advocates of using AI extensively in classrooms, they saw the physical classroom just as important as before AI for motivation and accountability. Completion rates for MOOCs, the open online courses that promised to democratise learning globally, sit below 13% on average, and in some cases as low as 7%. People enrolled because they wanted to learn but without any physical element, the commitment doesn't hold in the same way. Research from the University of Warwick looking at MOOC dropout patterns found that the absence of social pressure and accountability is one of the main factors.

Being physically present creates an obligation that being online doesn’t. You can close a tab and you can go quiet in a Telegram group. It’s much harder to simply disappear when you’re in a room.

Which brings me back to where I started. “Is Ethereum a network state?”

Probably not, at least not by the definition Balaji sets out. It has the economy and the permissionless tools. What it doesn’t have is the territory, and as the evidence suggests, that absence limits how far purely digital infrastructure can take you before it needs somewhere to land.

Balaji’s framing was “cloud first, land last, but not land never.” The research points to the “not land never” part being, really, a practical inevitability. At some point, if something is going to have real weight, it needs somewhere to be.

FAQs

Q: Is Ethereum a network state? A: By Balaji Srinivasan’s definition in The Network State, probably not. A network state requires a crypto economy, an aligned online community, and physical territory. Ethereum has the first two but no territorial claims. Evidence from DAO failures and online governance experiments suggests purely digital infrastructure consistently hits a ceiling without physical presence.

Q: What is the parallel polis and how does it relate to crypto? A: The parallel polis was coined by Czech dissident Václav Benda in 1978 to describe an alternative society with its own governance, economy, education, and information systems operating beneath a totalitarian state. Its seven pillars map closely onto what Web3 tools now enable. Projects like Paralelna Polis in Bratislava explicitly draw on this lineage, and they all have physical elements.

Q: Why did DAOs fail at governance? A: According to longtime Ethereum contributor Lane Rettig, DAOs largely failed because they tried to do everything online without any physical element. “If you’re serious about governance, you need to get the stakeholders physically in a room together,” he said. “It is the only way that humans build trust, period. No exception.” Elinor Ostrom’s Nobel Prize-winning research supports this by stating communities that interact face-to-face can adapt governance rules in ways remote coordination cannot replicate.

Q: Does a network state need physical territory? A: Yes, by Balaji Srinivasan’s own framing: “cloud first, land last, but not land never.” The definition in The Network State requires communities to crowdfund territory and eventually gain diplomatic recognition. MOOC completion rates averaging below 13%, DAO failures, and Ostrom’s governance research all point to physical presence being a practical necessity for sustained coordination, not merely a design preference.

Q: What does MOOC completion data reveal about digital governance? A: MOOC completion rates average below 13% and fall as low as 7% in some cases. Research from the University of Warwick identified the absence of social pressure and physical accountability as a primary dropout driver. The pattern mirrors DAO failures where people commit online, but without physical stakes, follow-through collapses. Being physically present creates an obligation that being online consistently fails to replicate.

Q: What did Nassim Taleb’s “Skin in the Game” say about remote governance? A: Taleb argues that accountability decreases the further you are from the consequences of your decisions both geographically and organisationally. DAOs, anonymous and remote by design, remove exactly this kind of stake. Without physical proximity to the consequences, the incentive to take governance seriously isn’t present in the same way.

The dark room, with reggaeton blaring and plastic wrapped booths momentarily threw me back to the Canarian discotecas of my youth, but I wasn’t there for dancing. I’d come to hear about the story of Bitcoin Beach, the El Salvadoran experiment in the coastal town of El Zonte, that sparked a revolution in the country’s approach to legal tender.

“Bitcoin beach was the only physical place that showed that bitcoin is more than a technology, more than an asset,” said Roman “Chimbera” Martinez, one of the pivotal figures in establishing the project. “Bitcoin is that tool that we were looking for.”

The project began in 2019 when an anonymous American donor wired Bitcoin to Mike Peterson, an American expat running an NGO in El Zonte. Rather than convert it to dollars, Peterson and Martinez decided to build something with it – a circular economy, the first of its kind, in a town of around 3,000 people where most had never held a bank account. The ambition, from the beginning, was not simply to introduce a new payment method but to prove that an entire economic ecosystem could run on it.

Young people were hired to do community work and paid in Bitcoin over the Lightning Network. Local businesses started accepting it. When the pandemic hit and the formal economy collapsed around them, the project expanded. People received wages in it, paid electricity bills with it, sent it abroad and received remittances through it, bypassing the Western Union fees that used to consume a significant chunk of what families sent home.

What happened next was the part no one had planned for. Jack Mallers, the founder of Strike, posted about El Zonte on social media. The posts reached Nayib Bukele, El Salvador’s newly elected president, who was already looking for ways to distinguish his young administration. In June 2021, Bukele announced at a Bitcoin conference in Miami that El Salvador would become the first country in the world to adopt Bitcoin as legal tender. The Bitcoin Law, passed by the Legislative Assembly that same month, mandated that all businesses accept it as payment, launched Chivo, a government-built wallet, and offered every Salvadoran citizen thirty dollars in Bitcoin to download it. Within weeks, a coastal town of 3,000 people had seeded a national experiment watched by the entire world.

The place of dreams

For the rest of the world, El Salvador’s adoption of bitcoin has been an economic success story, but for those on the ground, its biggest impact was on the mentality of the people who took it on.

Before the project, young Salvadorans grew up understanding, implicitly, that their future would have to be built somewhere else. The dream wasn’t to build something in El Salvador. The dream was to leave.

“We were seeing our kids dreaming to leave the country,” Chimbera said. “Thinking that our country could not provide opportunities and hope for us.”

El Salvador had long been defined by its problems of gang violence, poverty and an economy that offered limited formal employment to people without connections or capital. But Bitcoin Beach, in Chimbera’s telling, made staying feel like a choice rather than a failure. “For the first time, we’re dreaming,” he said. “For the first time we had hope.” He paused, then added, “I don’t want to live in the USA. I see my future in El Salvador. And 10 years ago, for a Salvadoran to say that, it would be impossible.”

The mechanism, he believes, is through opening a door, not necessarily making people walk through it. “A kid that has dreams, it’s a kid that’s opening a door,” he said. “I’m not saying it’s going to be possible. But that door is open.” And with it comes a restoration of the possibility to create something different.

I’m now back in Europe, the flashing lights, painted skies and windowless halls of Las Vegas left behind in a neon blur, but the idea of “creating dreams” has stuck with me in the days and long plane rides since.

Share

In 2009 political and cultural theorist Mark Fisher wrote about “capitalist realism”. In his specific framing, he writes about how capitalism has infused the collective thought so completely, that it’s impossible to imagine anything else. “It’s easier to imagine the end of the world than the end of capitalism,” he writes, and, as a Marxist, the book is written as a pillar of anti capitalist thought. But, whether you’re anti-capitalist or not, the idea of being unable to imagine more than what already exists, in my mind, prevails in modern society.

Focusing on the idea more abstractly, we can regard it as the pervasive feeling that it is easier to imagine the end of the world than the end of the current order, that the dominant culture has become exhausted, self-referential and incapable of producing anything genuinely new. The symptoms are visible in what Fisher described as a “slow cancellation of the future,” the inability to imagine alternatives that are not simply arrangements of things that already exist.

It’s seen in the engagement in two party political systems, despite not identifying with the policies of either, in the upholding of traditional institutions whose prevailing bureaucracy destroys the possibility of new treatments or tools coming to light, in the acceptance of and engagement in an attention economy that has been proven to reward polarisation and divisiveness.

Bitcoin Beach is an interesting case study because of its wider impact. The small scale experiment, focused on a tiny coastal town, proved its worth and evolved into an economy of bitcoin-infused, self sovereign businesses.

“It’s amazing how people in El Salvador are building on bitcoin,” said Juan Alejandro Diaz, founder of the Bitcoin Association in El Salvador. “In the fiat world we have upwork. In El Salvador we have Bitas. In the fiat world we have Uber. In El Salvador we have Bitride. In the fiat world we have the Apple store and we have the Bitcoin hardware store, and in El Salvador we have Bitcoin banks.”

Bitcoin Beach created a counter-narrative, one powerful enough to make people look at a place and think, maybe this is somewhere you can build a different future.

Proving another path

Bitcoin and digital assets are significant because they offered another path, not only to the flawed financial system bitcoin was built in light of, but other traditional institutions. For example, in my case, the first time Web3 really clicked was within the potential of NFTs to improve on the institution of Fine Art. A few years prior to my first foray into digital assets I had left the world of fine art, disillusioned by the abundance of gatekeepers and the lack of ownership artists had over their own success.

It seemed to me at the time that no matter how talented an artist was, if they didn’t fit into the prevailing zeitgeist of the art world, they were destined to fail. Instead of talent or a well thought out concept driving their ability to make a living from their work, it was who they knew, what they said publicly and whether it aligned with the select few at the top who made funding decisions.

Subscribe now

The eventual, return driven, NFT hype cycle aside (and whether you like the aesthetics of some of the collections or not) NFTs provided a mechanism for artists to directly find their niche of collectors, unfettered by the opinions of an elite.

Both NFTs and Bitcoin provided, at the beginning, small scale examples of systems that could be different. Their initial adoption was driven by people who bought into this promise and proved it had legs. Returning from Bitcoin 2026, I’ve seen that this potential is now influencing the approaches of larger institutions, some now set on enshrining self custody into law. After the initial hype cycle of NFTs making the assets more a meme than a tool, they still have a way to go. With the onset of AI generated imagery destroying traditional copyright protections, the use case may still be proven.

They demonstrated that the arrangement of things didn’t have to stay fixed, that the institutions we’d inherited weren’t the only possible ones and there was the potential to build something different. It could start in a surf town in El Salvador, or in a digital marketplace for art, and work its way outward.

What NFTs had the potential to do for the art world, and Bitcoin is doing for people’s approach to money, network states, startup cities and charter cities, may do for governance.

The petri dish

For those that have been following this Substack for a while you may recognise this argument from my foray into medieval charter cities a couple of months ago. In medieval France, the charter of Lorris proved the theory that giving the citizens of the town of Lorris more control over things like property rights, improved production of the town. It was so successful that the charter was replicated in other cities, creating a network of growing merchant towns that improved the economic standing of the country as a whole.

In my deep dive into AI’s impact on the classroom I touched on a similar subject, where small scale private schools were making headway on educational reform where previous attempts have failed. The founder of the Center of Educational Progress cautioned against expecting large institutions, and even people, to “just flip on a dime”, advocating for small scale experimentation that proved the worth of innovation instead of top down reform.

“You need to lay foundations for things, you need to demonstrate that things work, you need to be responsible and serious in the process of change versus just expecting that everyone will hop on board,” he said

In this week’s podcast I discussed the occurrence of governance experimentation in biotech and its impact in the wider sphere. Mac Davis, the founder and CEO of Mini Circle, was one of the first businesses to set up in Prospera, making use of the project’s innovation in governance to bring gene therapies based on 30 years of research but previously halted by FDA bureaucracy to market. Recent developments to biotech policy in America through the right to try bill in certain states has progressed this innovation further.

“Big ideas need to start small,” Chimbera told the crowd in that darkened Vegas nightclub. Bitcoin Beach started with an anonymous donation and two people with a theory that you could build a working economy from the ground up, in a place that had been told it couldn’t sustain one, using a tool that most of the world hadn’t heard of. They tried it, it worked, spread and gave the government a viable case study to point at when developing it further.

The existence of small scale experiments separates a hairbrained idea from a viable solution. It allows people to try an approach, and if it works, dream about its potential.

Share

Food for Thought

• If you want to find out more about the original Bitcoin circular economy project that preceded El Salvador’s legal tender law, head to the Bitcoin Beach website. Warning though, it may make a visit almost mandatory.

• In my research of capitalist realism, I came across this summary of Mark Fisher’s 2009 book which I found very interesting.

• If you want to go deeper into the idea of tradition and its impact on individual talent, TS Eliot’s original essay on the subject is a pretty good read.

• Mac Davis’s gene therapy company, Minicircle, ran its first clinical trials in Prospera before expanding to the Bahamas and beyond. Their story is a case study in what small jurisdictional experiments can unlock, and you can hear what he has to say about the subject in my latest podcast.

• I’ve linked to them in the body of today’s essay but my earlier essays, Petri Dishes and Classrooms in an age of AI follow the same logic in different sectors: small autonomous zones as proof of concept for the wider system.

Sounds like the stuff of sci-fi.

Mac Davis has been building the infrastructure to make that possible since 2019. He’s the co-founder and CEO of Mini Circle, the company pioneering mini-circle plasmid gene therapies. Follistatin, came first, which regulates body composition, muscle mass, chronic inflammation and epigenetic age, and they are now researching Klotho, which protects cognitive function, supports kidney health, and could be capable of something that sounds almost too good to be true - rolling back decades of mental fog in a matter of days.

In this episode, we dig into the therapies and why none of this was ever going to happen inside the US regulatory system. Biotech startups globally have voiced their concern over how many, potentially life saving, therapies are halted through the bureaucracy of regulators.

We talk about how Prospera became Mini Circle’s first research home and how they launched a clinical trial there in just eight months. But we also go beyond, into the global landscape for biotech regulation. Mac sees as a genuine turning point in 2026 a global restructuring of biotech regulation, from a Trump executive order on psychedelics signed just days before we spoke, to Saudi Arabia’s $400 million life extension pledge, Japan’s take on biotech, and how the Bahamas are constructing an entirely new regulatory framework from scratch.

Share

Topics Covered

• What Follistatin and Klotho gene therapies are and how they work.

• Mac’s first-person account of taking Klotho and what shifted.

• Why the FDA’s clinical trial framework made Mini Circle’s work structurally impossible in the US.

• Prospera as the first research home.

• Working with international governments: what’s on paper vs. what’s actually happening.

• The global “space race” for biotech regulation and why it’s accelerating now.

• Trump’s executive order on psychedelics and the right-to-try movement.

• Saudi Arabia, Japan, and the Bahamas making moves on life extension.

• How Mini Circle selects and screens clinical trial participants.

• Whether 2026 marks a real restructuring or just more noise.

Transcript

1:01 ISABELLE CASTRO

Hello and welcome to the Utopia in Beta podcast. I am here in Austin at the Mini Circle office with founder and CEO Mac Davis. And I’m very excited to talk to you about regulation around biotech and where we’re going, because you guys are like the pioneers of testing out new kinds of regulatory zones and doing some really interesting stuff with that kind of freedom. So maybe we can start with you introducing yourself, introducing Mini Circle, and we’ll go into all of the rest of that.

1:36 MAC DAVIS

Sure. My name is Mac Davis. We founded Mini Circle, me and my partner Walter Patterson, back in 2019, after I spent about five years of focused research trying to find what the ultimate biotech platform offering could be. When I was a kid I spent most of my time on the internet reading scientific papers and history. I was super interested in the history of information and information theory,the ways that people learned to encode, express, save, transcribe, edit, and share information. It started with language and evolved into uniform written language, money, address systems, and then eventually the internet, email. I got into crypto very early because it was the next obvious step in how humans transfer information and coordinate globally.

After getting into crypto I was like, okay, this is awesome, a really functional new kind of money, new contracts, more sovereignty with handling money, but I wanted something that was more impactful to personal health and personal experience. A lot of my life as a kid centered on dreams. My waking life was quite boring and uninteresting and I was mostly exploring the dream world. I thought: how great would it be if we could share dreams directly, almost like movies, that shared experience of what can often be a fantasy realm.

Subscribe now

So really the gene therapies kind of checked all the boxes for me. They allowed people to have more sovereignty and agency over their own health and genome. They allowed better access to medicine. And they allowed people, including myself, to explore what it’s like to take on new genes, to try these new genes on for a temporary period of time.

The Mini Circle program is basically this way to inject circular loops of DNA into your subcutaneous fat, similar to the way people do peptides or insulin, and then that gene is absorbed for a year and expresses in your body for a year. There are so many wonderful genes. Everybody has a different experience of the world because of their genes — everything is processed through that lens. So every time I’ve tried a new gene therapy, it’s been a different shift on life. It’s almost like changing your text into a different font. It’s still kind of the same thing, same identity, but everything has a different quality. And I think there are just so many awesome genes that people have that we can all benefit from.

5:40 ISABELLE CASTRO

And you’ve got two main therapies that are kind of out there — you’ve got Follistatin, which is more for cardiovascular and muscle health if I remember correctly, and then Klotho which is about to be launched more broadly.

5:58 MAC DAVIS

Yeah. So the first two gene therapies that we chose to test, both on ourselves, because I wanted to create these and have access to them and also get them out to the world and do clinical research were Follistatin and Klotho. These are two natural human genes. They’re in everybody. They’re part of human evolution and present in many animals.

Follistatin is widely known to regulate body composition, muscle mass, body fat, chronic inflammation, and epigenetic age. Klotho is more associated with kidney function, cognitive function, protecting the brain from aging, improving metabolism and fasting, and reducing calcium deposits in the body where they shouldn’t be — which is part of aging.

Both of these genes had been studied for at least 30 years in animals in laboratories around the world by great scientists. There were thousands and thousands of scientific papers about them. Researchers did population analyses to see what health outcomes were associated with different levels of these genes. They did decades of animal gene therapy experiments. Both genes were associated with lifespan extension in animals by 30 to even 40%. We pursued them because we wanted life extension gene therapies to be actually available to people everywhere — and these two seemed to be the safest, highest-impact ones to start with.

We started our first Follistatin clinical trial in Prospera in 2021, and since then we’ve completed five clinical trials — two of them on Klotho, many on Follistatin — and the effects have been astounding. Follistatin improves neuromuscular function, which is great for sarcopenia — people lose muscle and neuromuscular function with age, and it just makes you get stronger, less inflamed. Klotho has some pretty incredible impacts on cognition. We’ve seen significant IQ increases, speed of processing increases, fluid intelligence enhancement. Love Klotho. It should be available soon to more people, but we’re still researching it more deeply and working to provide a reliable care experience for everyone who takes it. Because the effects can be so psychologically significant — it can allow your lifestyle to change so much — I think having context and support available around that makes a lot of sense.

9:46 ISABELLE CASTRO

Can you go into a little bit of an example of what you mean when you say it could change people’s lives so much, and how you can support that?

9:56 MAC DAVIS

Yeah. When I personally first took Klotho gene therapy two and a half years ago, it was definitely one of the most transformative experiences of my life. First day, I remember waking up and going into my kitchen and sitting in the hallway and just seeing everything in my house extremely clearly. Very wide vision. I could see every object I knew in its fullness, and also the space in between everything.

I got up to my coffee machine and made my espresso, which I had been doing every day before I took Klotho — and I was just hyper aware of every step, my mind going through that habitual process. Everything had gotten a new perspective. And I felt like at every step of that habit, I could just make a decision to change it. I could just shift. In the study of psychedelics and the brain, there’s this concept called the default mode network — it’s kind of the way your brain assumes things are, and the way your brain assumes things should be. It actually costs your brain and mind energy to make a change from your ingrained habits. But after taking Klotho, it basically cost nothing to just shift any habit. I was hyper aware of the whole process the whole time.

Share

That day I also listened to some music on my speaker system and it was the most astonishing thing. It was like I had never heard music before and this was my first time. I remember I had some early experiences with water — I hate swimming, I hate large crowds of people, I hate it when it’s hot outside. And I thought, everything’s new, I’m going to go try swimming in the local pool at Barton Springs that all my friends love and I always hate. And I went in there and I just had a great time. No pre-expected aversion. I was literally able to do it without reacting the way I had always automatically reacted because of early childhood trauma responses.

We’ve recorded at least 60 hours of video testimony of people going through the Klotho experience from beginning to end. I’ve heard these types of stories from many people — just astonishment waking up with perfect witnessing clarity, a lot of reduction in brain fog, feeling like rolling back your mind’s aging by 20 or more years. That type of opportunity needs to be respected and used consciously. It’s not just: boom, here’s your shot, now you’re healthier. It’s totally unique to each person — what this means to you, what you can get out of it. I think eventually Klotho is a fantastic gene to give to an entire population, because it improves IQ, improves cognitive function, gives people a freshness, protects against aging. Especially in this era of AI — with people afraid of humans losing relevancy — I think it’s really critical that our brains are adapted to new circumstances.

15:31 ISABELLE CASTRO

I remember when we first talked about Klotho here and it always sounded amazing. Was there a specific moment when you realized you wouldn’t be able to start development for these therapies within the US system?

15:53 MAC DAVIS

There was a specific moment. After we realized the plasmid platform was working — the plasmid is like a template vector, a cassette, a packet of information you can put almost anything into — we understood that if the plasmid worked, that implied thousands of these gene therapies, maybe even more, would be expected to work.

The way that drug design and FDA regulation has always been handled is that every drug is often treated as a completely new drug. The marginal cost of running trials doesn’t go down. Every series of clinical trials will cost millions, even billions of dollars over time, and none of that can carry over to the next one. But because this is an iterative template platform and we were using bioidentical human genes that had already been through many clinical trials themselves, or had been studied in animals for 20 or 30 years — there was already a lot of background research done, a lot of safety already implied. I just realized we had to find some other option. The technology was too advanced and too real to potentially be left to die in research, as it had been for decades. The same life extension gene therapies had been sitting in freezers in academia for decades while everyone was afraid to try them.

I looked for another option. I called a couple of friends who eventually introduced me to Prospera. We had also been exploring Ukraine, New Zealand, Costa Rica. But in Ukraine the war started; in New Zealand, COVID lockdowns. When I arrived to Prospera it was just amazing. My first day there there were economists from the EU, the guy who designed the Estonian e-residency program, the chief strategist of the Dubai Financial Zone. Just every person I met was astonishing and on the cutting edge. The vision was solid. And so I chose to begin our initial research there.

We also met up with the GRM clinic, the stem cell clinic that had been there for over 10 years, with Dr. Glenn Terry. Our experience with them was fantastic. Still probably the best clinic we’ve worked with over the last nine years.

20:08 ISABELLE CASTRO

And you managed to do a clinical trial just eight months after arriving. What was that timeline like in practice?

20:35 MAC DAVIS

I think I first showed up in March of 2021, something like that. I had already created my company. I got an office there, got connected with the clinic, went through the insurance process, went through the regulatory process, recruited our trial, and designed the IRB package.

If you’re doing clinical trials, you always need to have a set of documents — a package that in detail describes what the study is, what the product is, why it has value, why the study should occur, who the physicians are, where and when the study begins and ends. Then you get approval from what’s called an Institutional Review Board — essentially a group of real adults, not all doctors, who come from multiple professions and can look at the research and ethically agree that this is a good idea. We prepared our first IRB package, got it approved by an international IRB, and started the trial. Probably within eight months.

We had some of the coolest people I’ve met in the longevity space come down and try this new gene therapy that they had been wanting to try for many years — in such an adventurous jurisdiction. A tropical island, a crazy hotel, this clinic. A lot of the first trial participants were really in it for the adventure, and they were so committed to longevity and health that they had basically done everything they could for 20 or 30 years already. This was kind of the next thing.

And I think that adventure is also part of the consent filtering process. If you go through a multi-month process to evaluate an experimental gene therapy, fly down, spend days as part of this adventurous research group in a special economic zone on an island off the coast of Honduras it’s hard to say you didn’t really commit to this or didn’t understand it was something new. It’s not like going down to Walgreens with your doctor who says, “Yeah, you have this, you have to take this. Okay, I’ll take it.” And you don’t understand what you’re doing.

24:28 ISABELLE CASTRO

Since then you’ve gone to some other jurisdictions that are testing out new ways to regulate around biotech. Tell me about those, because there’s much less written about it online.

24:37 MAC DAVIS

Most of what’s going on in this type of space with jurisdictions changing is not online. Even Prospera you kind of had to be there and talking directly with the regulators, the government, the policymakers, the people who feel like stakeholders in the jurisdiction, to know what was really going on. Every place has totally different nuances.

There’s an inherent nature to it: you have rules that are on paper, codified, maybe very old, certified by the legislature at such and such time and then there’s the reality of social approval. Anytime you enter a kingdom and want to do something novel there, you kind of need the approval of the king, or the kingdom. They have to know what’s going on. They have to be into it. It comes down to that in a lot of ways, even in a fully democratic society with all these bureaus.

So in terms of new things going on: in the US there’s the right-to-try movement, which has made a ton of progress on both passing laws and actually treating patients, and getting into the rhythm of realizing: okay, there are so many valuable experimental therapies, and there are ethical reasons people should have access to them. You can’t tell somebody with a terminal diagnosis they can’t try something that might cure them just because it hasn’t been tested on tens of thousands of people first. We saw this with HIV — it was totally new, deadly, spread rapidly, and there were tons of intelligent, well-informed people willing to try new experimental HIV therapies, both to help themselves and to help their community. Different states are passing different right-to-try laws, and the field is just changing so rapidly that even what is official, people don’t necessarily know.

28:20 MAC DAVIS

Just four or five days ago, Donald Trump signed an executive order further pushing for the right to try specifically around psychedelics, including ibogaine, for mental illness and for the VA for veterans. It opened up the possibility of psychedelic therapies for people with mental illnesses, and a priority review and approval pathway, and a directive for the DEA to immediately consider unscheduling any psychedelic drug that’s been through a Phase 3 trial. That’s huge. We had actually just hosted an event here in Austin with the same people who ended up in the White House five days ago, behind Trump as he signed the order which was a funny thing to see.

The Middle East is making a lot of moves, though obviously the current geopolitical situation affects everything. Saudi Arabia made a big announcement on longevity something like a $400 million pledge to invest in life extension technologies. Japan is also making a lot of moves. They have a new president, a new political party that’s aligned with the president, and they are serious about addressing their demographic decline. Japan has an aging population, a low fertility rate — it’s really an existential issue for them to do something about biotech, aging, and infertility.

Over the last 10 years, Japan’s been making improvements toward allowing expedited and earlier access to regenerative medicine. But I think only this year is there more of a real turning point for the first time, induced pluripotent stem cell therapies were approved this February, two of those therapies that had been under development for over 10 years. And it’s worth noting: Dr. Yamanaka and Dr. Kuro two of the greatest discoveries ever made in medical aging science were made by two currently living Japanese scientists. Yamanaka discovered basically the mechanisms cells use to reverse aging. Kuro discovered Klotho in 1991 and only published the paper on it around 1995 to 1997. Ironically, Japan which has been so conservative about allowing new medical research to reach its people also made two of the greatest discoveries in medical aging, and they are at some of the greatest risk of losing their culture to demographic decline.

The Bahamas, too, they had a very successful regulatory initiative around stem cells that started 10 or 15 years ago. They created a novel framework to allow stem cells to be researched and provided to people there. It was great for their economy, great for everything. No weird scandals. Everyone was happy with it. So they decided to take it further and create a new regulatory framework to allow not just standard stem cell therapy but a wider variety of new biologics for longevity, health span, life extension, and advanced regenerative medicine. We recently were commercially approved by the Ministry of Health to offer the Follistatin gene therapy there in the Bahamas. That just happened this year.

Everything is really changing. When I was a kid I thought the government never changes, nothing ever changes, the only thing that’s changed is the internet. But the internet has finally started to change everything else. People have gotten more access to personal health information there’s so much great information about nutrition, exercise, and peptides that people never had before. And I think that, because everyone is getting more informed about personal health and longevity, regulations are basically loosening up and giving way toward allowing people to get more access to both what they want and what we can now actually invent. Scientific discoveries shouldn’t just sit in books and labs. They need to get into service so everybody can absorb the benefits.

38:03 ISABELLE CASTRO

It sounds like there’s almost a space race for biotech regulation. Do you think that’s come about because people are more informed and putting pressure on regulators? Why is it happening now?

38:24 MAC DAVIS

I think regulators feel safer because people are more informed the power asymmetry isn’t as intense as the old dynamic of “these people are predators and these people are helpless.” But what’s really going on is that jurisdictions are competing, because they can see the value of these therapies is there. Smart people in power have already identified where this is going and they want that economic activity in their country, their jurisdiction, their state. In Montana, New Hampshire this has been what it’s like: the economic value and health value of these new inventions is so compelling that they want to vitalize their state and bring that innovation. That’s what I’ve heard from people in governance in those states. That’s what I’ve heard in Japan. That’s what I’ve heard in the Middle East, in the Bahamas. And I think that’s what’s going on with the FDA as well they know that other countries, including China, are making regulatory changes to try to get a step up on medical innovation. The US of all places doesn’t want to get left behind.

We’ve also had activists work hard for many decades, basically since 2000, to provide credibility and reality to the concept of science allowing changes to the very nature of aging. People like Aubrey de Grey scientists were laughing him out of the room for years. Even Bryan Johnson when he started, people were like, “Who is this weird guy?” But more and more every year, people are starting to realize this is real, and we have the opportunity to genuinely improve health and potentially extend lifespan. The space race comparison is a good one. Regulations have changed around cannabis, psychedelics, crypto. The world has changed. So yes.

43:08 ISABELLE CASTRO

How much of an impact do you think the regulatory bottleneck has had on the progression of longevity not even just research, but the impact of that research on our own lifespans?

43:26 MAC DAVIS

I think it’s basically 100%.

The science has been there for so long for so many of these things. Having a single authority that is difficult to navigate for the average inventor, that requires often a decade or more of experiments and where the consent model has been more regulator-consent-lean than individual-consent-lean that’s a problem. There’s an assumption embedded in a lot of historical clinical trials that’s fundamentally wrong: you want to recruit people who know they’re in a trial, but how well do they really understand the drug they’re taking? So many aspects of it are desired to be totally objective. The patient in the trial is just an object, the same as every other one. But that’s inherently against the nature of medicine. Medicine is so much in the subtlety in knowing the health history, the lifestyle of the patient, the relationship between patient and provider.

In the industrial pharma, post-WWII paradigm it was very much: this is a factory, people are objects, this is a scientific study. I think that results in a lot missing from an ethical and care-providing standpoint. Providing care should start at the very beginning of the invention process and be kept all the way through.

The way medicine was done before modern clinical trials was actually great. Physicians would try new things they had the freedom to do that and they would report results through medical journals. Try it on one person, do a case report, other doctors evaluate it, gradually expand it through many people and continue reporting results. It was organic. There was no Phase 1, Phase 2, Phase 3. IP and copyright are relatively new inventions in history. Einstein didn’t create his theories of relativity in order to IP-protect them. Bach created all his works before copyright was ever thought of. Humans have been creative for millennia before these recent legal structures were created. The history of medicine, ethnobotany and healing work, was legitimate in many ways, and science is great, but there’s a lot more to it.

47:48 ISABELLE CASTRO

Critics of right-to-try often see the FDA as the standard that ensures safety. How do you maintain standards while having something more organic and peer-reviewed?

48:11 MAC DAVIS

I really believe in naming the things and naming the different standards. There can be more than one standard. Jim O’Neill has advocated for decades for the concept of: okay, this drug has passed a Phase 1 or Phase 2 trial in safety but maybe the efficacy hasn’t been formally reviewed by the FDA yet. If we’re optimizing for safety, you could theoretically have a label or a website where people can look up: “The FDA has done this type of safety review on this dosage and this is approved the efficacy review is not yet complete.” You could just look it up.

Safety compared to what, anyway? Is this safe for everybody? Is this safe for people with certain diseases? I really believe it’s possible to communicate that to people in a much higher resolution way than a simple binary “yes, completely safe and effective” or “no, not safe and effective.” We got into that type of trouble with the COVID vaccine it was emergency approved in the shortest time ever for a therapy of that type, given infinite “yes” on safety and efficacy, and now not everyone totally believes it was infinitely safe and efficacious for every person. That’s the nature of that approval process it leaves out so much information. No drug is completely safe and effective for everybody in every case. It just depends on the dose and the person.

I also think about how other regulated industries handle this, food products, for instance. There are nonprofit entities that certify products as having passed different types of safety tests. There’s halal, there are religious certifications, biodynamic certifications where auditors go out to farms and check if they’re adhering to standards of growing food. Those entities even non-governmental ones are generally very highly respected. It is possible to do this in more than one way, and having a plurality of regulatory approaches where you have different levels of certification for different things is great. You shouldn’t have to earn one single certification that blanket-approves everything.

A striking example of how limiting the current model is: the first drug approved for female libido enhancement, flibanserin, had already completed Phase 3 trials years before. The FDA said no, this isn’t a national priority, this isn’t a real disease. That company had to go on a multi-year propaganda campaign, appealing to feminist ethics and lobbying politicians, in order to get another review. And then the FDA approved it without any additional science, they had already completed everything. When you’re a startup founder burning cash every month, the FDA essentially decides whether your company lives or dies based on the ambiguity of when approval might come. And that approval has a lot to do with whether the FDA has decided your condition is even a disease at all. When there’s a single authority that controls the ontology of disease — creates boundaries, names things, defines what is and isn’t an illness — you’re going to have an inherently limited view of reality. And wherever people don’t fit into that structure, they go suffering, potentially for their whole life.

57:25 ISABELLE CASTRO

I’m curious, when you go into your trials, how do you actually do the process of accepting applicants?

57:42 MAC DAVIS

Trial recruitment can be one of the most interesting parts of the process. A lot of it comes down to determining whether the participant is a good research candidate and that includes several things.

First: are they sovereign? Familiar with their own health? Able to respond effectively to anything that goes wrong, take responsibility, call a doctor, stay on top of their own health? Then there’s how that connects to personal stability. Maybe the person is really attracted to trying this new therapy, but they’re actually schizophrenic or prone to anxiety attacks, and some part of them isn’t fully consenting and maybe they’re even trying to hide that. You kind of have to personally connect with the person in order to tell if their statement of consent is really full consent.

If it’s a clinical trial, the agreement is really for them to provide data it’s not just to take the therapy. People need to follow up with their data. That means showing up on a schedule, going through various batteries of testing, blood draws, tests of behavior and intelligence, urine samples, all kinds of things. We might have people in for six visits, and that cuts into the time of day. Sometimes people just want the therapy and want to get out of there. High-agency, wealthy people with busy lives often know everything about the trial and have full consent, but genuinely can’t commit six plus days to a clinical testing procedure they just want to pay for the therapy and go back to their lives.

Lastly, I’m really looking for people who are sensitive to changes in their own body and mind people with experience in biohacking, maybe even psychedelics or performance-enhancing drugs. People who have a lot of self-awareness and can articulate what has changed. Not everybody is dropped into their body and sensitive enough to their own psychology to easily name: “Oh, this has changed” or “I’m noticing this.”

1:02:02 ISABELLE CASTRO

So looking at the current landscape, the space race, the regulatory innovation, all of that, do you think we’re on the brink of a genuine restructuring of how the world approaches biotech innovation? Or is this still going to be a situation where jurisdictions compete and not everyone gets access?

1:02:29 MAC DAVIS

I think this is the year where I’m seeing restructuring. I’ve always been very skeptical maybe more than most of any type of regulatory restructuring actually happening. Usually governments or people say they’re willing to do it, but this is the first year I feel like I’m seeing things actually change and allow legitimately new, advanced biotech therapies to be welcomed, explored, and made available to people. Not just seen as research artifacts in sterile NIH-funded laboratories.

1:03:24 ISABELLE CASTRO

How much of an impact do you think was made by smaller-scale experiments like Prospera or the really tiny charter city approaches to regulation?

1:03:37 MAC DAVIS

I think the choices made by Prospera and probably by Mini Circle as well can compound. What I was seeing before was that no country, especially not a bigger country, but also not smaller countries, wants to be the first to try something no one else has tried before. But I think people have taken note of both new innovations occurring in the traditional regulatory system like GLP-1 agonists, semaglutide almost becoming a longevity drug and then these more radical experiments, like what Mini Circle did in Prospera, that actually turned out really great. It wasn’t a catastrophe. People had greater health outcomes. Bryan Johnson did it, made a video about it, and more and more people have been trying it and saying, “Oh, this is working. This is okay. Things can be good.”

My challenge to the world has always been: okay, so the world is like this but can you imagine how much better it could be? I think smaller countries being willing to try things, even at a very nascent level, helps governments and people who work in government feel more comfortable. It demonstrates that experiments have happened and this can work.

There’s also been a lot of work done in private by governments and regulators particularly in the UAE and the Bahamas on what would a regulatory system look like if we designed it from scratch today? The FDA was designed a very long time ago and is very big, and has to incrementally adapt. These newer frameworks were built with a fresh perspective. A lot of that background work is only now becoming certified the orders are being signed. And then AI, I think AI has really forced a lot of people to say: okay, things are genuinely changing. There’s this old philosophical concept maybe Diderot about certain times where society melts and becomes plastic, and a lot of changes can happen, and everyone’s kind of in on it, all seeing: it’s time to redo things. I think we have entered that type of timeline. I think AI has a lot to do with it. And the internet has been around long enough that it’s not just for nerds anymore. It’s part of the infrastructure. And when you get new infrastructure that’s accessible to everybody, you can reach the next level of civilization. That’s what’s happening.

1:08:38 ISABELLE CASTRO

Okay. That’s a very optimistic place to end it, and I’m going to leave it there because it’s a great note. Thank you so much for listening to this episode. Subscribe, like, share, and go look at Mini Circle’s website and everything they’re doing. Bye.

FAQ

Q: What is Mini Circle and what gene therapies do they offer? A: Mini Circle is a biotech company founded by Mac Davis in 2019 that delivers gene therapies via circular DNA loops injected subcutaneously — similar to how people take peptides or insulin. The gene expresses in the body for roughly a year. Their first two therapies are Follistatin, which regulates muscle mass, body fat, chronic inflammation, and epigenetic age, and Klotho, which supports cognitive function, kidney health, and protection against brain aging.

Q: Why did Mini Circle have to leave the US to run clinical trials? A: The FDA treats each new drug as a unique entity, meaning every clinical trial series costs millions to billions of dollars with no carry-over to the next. Mini Circle’s plasmid platform is a template — the same delivery mechanism for potentially thousands of gene therapies — but the FDA’s structure can’t accommodate that iterative model. Mac Davis concluded the technology would die in regulatory limbo if they stayed in the US, where life-extension gene therapies had already been “sitting in freezers in academia for decades.”

Q: What does Klotho gene therapy actually feel like? A: Mac Davis describes his first day on Klotho as one of the most transformative experiences of his life — extreme visual clarity, hyper-awareness of habitual thought patterns, and a near-total dissolution of what neuroscience calls the default mode network’s resistance to change. He found himself enjoying swimming in a crowded outdoor pool, something he had always hated due to childhood trauma responses. Mini Circle has recorded over 60 hours of participant testimony describing similar effects, including significant IQ increases, faster processing speed, and the sensation of rolling back cognitive aging by 20 or more years.

Q: Which countries are leading the global biotech regulatory shift in 2026? A: According to Mac Davis, 2026 marks the first year he’s seen genuine regulatory restructuring rather than just political promises. The Bahamas has commercially approved Mini Circle’s Follistatin therapy and is building a new regulatory framework for advanced biologics from scratch. Japan approved its first induced pluripotent stem cell therapies in February 2026. Saudi Arabia pledged approximately $400 million toward life extension technologies. In the US, a Trump executive order signed in early 2026 pushed right-to-try pathways specifically for psychedelic therapies, with DEA scheduling reviews to follow.

Q: What’s wrong with the FDA’s binary approval model for new therapies? A: Mac Davis argues that the FDA’s binary “safe and effective / not safe and effective” model collapses too much information into too simple a verdict — and that no drug is safe and effective for every person in every dose. He proposes a tiered system in which safety approvals, efficacy approvals, and population-specific outcomes are all separately labelled and publicly searchable, similar to how non-governmental bodies certify food products for different standards. The COVID vaccine approval, he argues, demonstrated the cost of the binary model: an emergency approval given infinite credibility that ultimately not everyone believed applied to them.

Q: How does regulatory competition between jurisdictions affect longevity research? A: Mac Davis estimates that regulatory bottlenecking is responsible for “basically 100%” of the delay between longevity science being ready and longevity treatments reaching people. He argues that jurisdictions are now competing economically — not just ethically — for biotech activity, because governments in Japan, the Middle East, the Bahamas, and US states like Montana and New Hampshire recognise the jobs, investment, and health value of hosting this innovation. Prospera in Honduras was an early proof of concept: Mini Circle launched a clinical trial there eight months after arriving, in a jurisdiction willing to approve a novel gene therapy that no US institution would touch.

Patri Friedman has spent the better part of two decades figuring this out, first through the Seasteading Institute, and now through Pronomos Capital, the only charter city-focused investment fund in existence.

In this episode, we dig into Alpha City, the first project Pronomos has incubated directly, targeting different locations in Africa, and what Patri has learned from watching early charter city projects like Prospera struggle with the “cold start problem.” We also talk about the critical sequencing mistake that kills these projects before they begin, why jobs, not laws, are what actually move people, and Patri’s upcoming online course ‘How to Start Your Own Country,’ which might be the most ambitious syllabus ever written.

Share

Topics Covered

• What it takes to create a new country and why knowing the path doesn’t make it easy

• Alpha City: Pronomos Capital’s incubated project

• Securing government deals before buying land

• Charter cities as ‘venture capital returns at the scale of real estate’

• The cold start problem and why even saying ‘new city’ is a paradox

• The real reasons people move and what charter cities must get right

• Learning from Prospera and how to bootstrap an economic engine by hand

• Applying Silicon Valley’s ‘talk to your customers’ principle to nation-building

• The bottleneck shift: from willing governments to qualified founders

• Patri’s upcoming course: ‘How to Start Your Own Country’

Subscribe now

Transcript

1:01 ISABELLE CASTRO

Welcome to the Utopia in Beta podcast. I am joined today by one of the forefathers in network states, Patri Friedman, who needs almost no introduction. You were the founder of the Seasteading Institute and also have been a prolific founder of a lot of the network state projects. So I’m looking forward to getting into the funding side, but first I want to ask you — in all of your years that you’ve been engaging in the movement, have you distilled it into what it takes to create a country?

1:44 PATRI FRIEDMAN

Well, starting off with a big question. Yeah, I think so. Knowing the path and the milestones and the ingredients is super helpful, but it’s still incredibly hard to do. I feel like I have a sense for what’s needed, but it’s still an epic challenge.

2:07 ISABELLE CASTRO

And right now, your main focus is on Alpha City. Maybe we can start with you outlining the idea around Alpha City and we’ll go from there.

2:18 PATRI FRIEDMAN

Yeah. So I run Pronomos Capital, which is the only charter city focused investment fund. We fund cities that write their own laws, that have formal regulatory autonomy. We had a five-year investment period from 2020 through 2024, and so now one of my focuses is helping our portfolio grow. This company, Alpha City, is the only one that we incubated. My business partner Bradford Cross, who helped me run Pronomos for years, is the founder and CEO and he’s a really experienced serial entrepreneur, and that’s been a key ingredient because it’s so hard that to have a good chance, it takes a lot of experience and badasserie. So Alpha City is initially focusing on Portuguese-speaking countries in Africa because I have excellent connections there, and also talking to some other countries in Africa. One thing that’s different about the Alpha approach is that we’re trying to do it at scale — we’re talking with multiple governments, have with four of them, and are close to land deals with a couple of them. One thing people find in the space is that governments are just slow and unreliable. It’s like enterprise sales, but governments are like a giant business compared to a small company. They’re just huge, so it’s slow. If you just work with one country, you’re just going to be stuck. So we’re working with a bunch of countries in order to have multiple irons in the fire, as well as hoping to get multiple deals for land, pass legislation, and build in multiple places at once using the same formula, the same regulatory system. Some of the target industries will be the same, like agriculture in Africa — upgrading the quality of their agriculture is just huge everywhere. And then there’s more specific industries, like Kenya has the world’s largest amount of untapped geothermal energy that you can tap into and build AI data centers on.

4:57 ISABELLE CASTRO

Okay. So you’ve already kind of created these connections within these governments, but for someone who is new and they want to found a country — where do they start? Is it in the funding? It sounds like it’s not from the land, even though a lot of people do focus on the perfect plot of land. I think you just did a post about that.

5:19 PATRI FRIEDMAN

Yeah. My friend Devon Zoo, who worked with me a bit at Pronomos in the early years, was working on this project called Esmeralda in Northern California. In her kind of experience and research she said the value of land is the chance that somebody will let you build on it. Sure, things like the views and the water and the location — those things matter, but ultimately that land is useless if you can’t build on it. In a lot of parts of the world, California especially, governments make it really hard to build — zoning and permitting and all these things, and environmental review. So the thing that matters the most about a piece of land in California is not the land, it’s whether someone will let you build there. Bology has this great line describing charter cities as ‘the returns of venture capital at the scale of real estate,’ because it really is a hybrid between those industries. I feel strongly that can be true if you do it right, but if you buy your land before you have a solid deal with the government — not an MOU, I mean a joint venture legislation — then you’re having these huge capital costs of buying land come before the project is derisked. My line is that you have the capital cost of real estate and the failure rate of venture capital. That little thing about the sequencing is critical.

7:04 ISABELLE CASTRO

I want to go back to that — ‘returns of venture capital’ — because I feel like network states and charter cities operate in this weird kind of in-between place where they need the funding and they’re getting quite a lot of funding from kind of venture capital, but these are long-term projects which may not get the returns at the speed venture capital is accustomed to. How do you navigate that? I know Cabin, for example, that wasn’t a good outcome for them ultimately. How do you approach it?

7:49 PATRI FRIEDMAN

I think of it as a hybrid industry. You can potentially get much higher return multiples than most real estate. If a project has physical assets and then it fails, it can still return something — maybe even make a little bit of money. But it’s never going to grow as fast as a software company because you’ve got to actually move dirt to expand. I just expect that these projects will grow more slowly and that means the IRR compared to venture capital is lower because your returns are stretched out over more years. But I just really think that the long run is not the problem. If you can create a physical community with an economy that’s growing, there are all kinds of ways to finance that. You start looking more like a city and you can use debt, and tap into the world’s capital markets for real estate, which is a much larger asset class than technology, much much larger than venture capital. Most of our world’s value is in real estate. I think the hard part is getting to profitability. That’s incredibly difficult and almost no one is doing it or close in this brand new space. The attention needs to go on getting to profitability and a growing occupied community, and then bringing in more capital at that point is easy. Everybody loves non-dilutive capital — if you’re an angel investor in a company and they get this half-million dollar nonprofit grant to use their technology, you’re like ‘yay.’ And raising debt to continue building out the city and its infrastructure — which you can do because you’re building hard assets and demonstrating that there’s demand for it — is non-dilutive capital that you’re going to get later. So it’s true that these projects will grow more slowly and the most successful ones will probably not have Facebook or Google levels of return, but you’re putting in money when capital requirements are low at the early stage, especially before they need to buy land or build — and there’s potential for it to keep growing for decades with other people’s non-dilutive money.

10:51 ISABELLE CASTRO

Going back to what you said about not that many reaching profitability — well first, there’s hardly any that have launched.

11:05 PATRI FRIEDMAN

True. I think Sealand — which is not a charter city — is profitable in its own way. But there’s very little in the space that is. It’s early stage. We’re all struggling to figure this out and get it done. To do a charter city, you’ve got to get government to pass national legislation, or maybe a province. In the US, most of the money and the laws are at the federal level, but there are other countries where the provinces have a lot more autonomy, and of course that makes it easier. But it’s really hard to get to actually having a place that can write its own laws that people live in — and you’re not going to be profitable until after then.

11:50 ISABELLE CASTRO

What do you think are some of the stumbling blocks towards profitability? I know in the past you’ve talked to me about critical mass of community. Is it that, or is it something more than that?

12:11 PATRI FRIEDMAN

I don’t think there’s a stumbling block to profitability. I think there’s a stumbling block to actually building a real charter city. The real estate model is pretty clear — you can collect taxes and fees like a government and collect rents, capital appreciation like a real estate person — but it only works if you can make a place that has an economy and a people, and that’s hard. In survey data, the two biggest reasons people move — both in the developed and developing world — number one, by far: jobs. I feel like projects in the space are doing too little with bootstrapping the economy. There’s this mistake people make where one of the theoretical drivers of all of this is all the data from development economics that societies with less corruption, better laws, lower taxes, British common law-based, have higher rates of GDP growth. If you can rezone land with better laws and institutions, people on that land will get richer faster — that’s the Singapore story, that’s Deng Xiaoping bringing in more capitalism and lifting half a billion people out of poverty. I believe in this, but it’s only something that happens once the economic engine has started turning over. Once it’s kicked off and running, sure — I think these places will have this higher growth rate — but it doesn’t magically happen just because you rezone land with better laws. You have to find the people to live there, figure out the industries that will make jobs, talk to the anchor tenants, build the core infrastructure, figure out your minimum viable first build. As a founder you have to be doing this business development by hand — figuring out your industry clusters, what are the skills and resources the people here have, what kind of people your vision of a society attracts, what kind of things do they do. You have to build that economic engine by hand and have it start achieving some scale before any of those long-term effects come along. So one is jobs — I think there needs to be a lot more focus on that.

Number two reason people move is family reunification. If you’re going to some new society, you’re probably moving away from most of your family members. That’s a negative. That’s why, with Alpha City, we’ll seek to attract some really aligned nomads and remote workers — especially people working with our crypto partners that are going to use the new financial center that we’re building. But a lot of it is building for the locals, right? They can move into your zone and not be moving too far away from their family. Or the reverse diaspora — in countries like El Salvador or India, people coming back because their country’s gotten better. Those people bring business experience and capital and knowledge of a society that functions better, and moving back is going with family reunification.

It’s hard to get people to move to a new place without very many people. Prospera has really struggled with this — it was a while before they were able to build their first residential tower, it took a while to sell out, and it’s now mostly occupied most of the time, and they’re building some more. But it’s hard to get people to live there. It’s easy to get people to visit, it’s not that hard to get people to come check it out for a month or two, but getting people to settle down in a new small place is hard. We call it the cold start problem. I’ve been kind of sadly joking lately that even to say ‘new city’ is a paradox, because cities depend on network effects and new things don’t have network effects.

17:13 ISABELLE CASTRO

Learning from Prospera and taking the learnings that you just outlined — how are you applying them to Alpha City? In terms of the economic engine and all of those things.

17:38 PATRI FRIEDMAN

A lot of it is just about putting that first. If you have an economic engine and nothing else, you have sustainability and you can figure out the other things. If you don’t have an economic engine, you’re burning money and you’re stuck trying to fundraise for it and you’re on a clock. A lot of what I do is just tell people to do the simple basic things. In the Silicon Valley startup world we have this book The Four Steps to the Epiphany and this whole approach where you don’t build some giant thing and then try to sell it to people. You try to figure out your potential customers, figure out the smallest things you can build with them, talk to them and engage them. For example, if you’re having a popup that you hope will attract people to move someplace, talk to as many people as you can afterwards — beforehand like ‘why are you interested, what are you looking at,’ and talk to them afterwards: ‘are you more or less likely to move now, what were your decision points?’ It’s just this basic: talk to your customers. The charter city space just appeals to a lot of founders who are not from startup culture, right? And this kind of insight — which has become absolutely foundational to the startup world — is missing. I think it’s kind of less intuitive, right? It’s intuitive to be like ‘let me work on my thing and make my thing and make it as perfect as possible and then show it to people and they’ll love it.’ But in practice, you have to engage with your customers.

19:29 ISABELLE CASTRO

I imagine you get a lot of people approaching you kind of with their ideas for new societies. Is there a prevalent trend in what they’re trying to do? Not the ones that you actually invest in — just in general.

19:43 PATRI FRIEDMAN

I think the biggest set is people working on projects that aren’t about any legal autonomy. I’m kind of calling this the startup villages space, but legal autonomy is super difficult — you shouldn’t do it unless that’s really what you’re into. I feel like there are a lot of things about how we live together. I’m an intentional community guy — I started two of them in the Bay Area in like 2005, lived in intentional communities for almost half my adult life. I think more co-housing communities built around shared values and a shared idea of what a healthy life looks like — that’s great and we should have lots of them. But my north star is the legal autonomy and there aren’t funders in the space for these startup villages. So they’ve been coming to me because I’m the funder of this adjacent space, and yeah, it’s a big funding gap. Normal real estate people are like ‘wait, there’s a crypto component’ — even for co-living communities, there are very limited numbers of lenders. The space being new, there’s this huge funding gap. I’m hoping that over the next couple years I can start additional funds with these other focuses. I’ve got a couple people I’ve been talking to about a startup villages fund, because there are a lot of great projects that need funding and there are big gaps.

21:45 ISABELLE CASTRO

No, that’s really interesting. I didn’t realize there’d be a gap there — because in my mind, real estate development and all of that funding structure is already quite established.

21:57 PATRI FRIEDMAN

It reads as a new, weird type of project. And maybe if somebody who has experience raising money for real estate development is able to make it look normal — that’s great. But for a lot of these projects, there’s some component, some part of the vision, that’s new and different. And it’s just a lot harder to find a vanilla real estate investor who understands this new space.

22:45 ISABELLE CASTRO

It almost feels like a kind of impossible situation at this point — and I’m not putting a dampener on things — but a startup village is a lot more attainable. I can understand why people are going for that more, because talking to governments just seems like a big thing, and I don’t think people really know where to start with that. So that actually brings me into your course that you’re developing, which is my understanding is going to prepare founders who do want to engage in this whole new governance system — a kind of pathway into understanding what that means. Tell me about it.

23:39 PATRI FRIEDMAN

Yeah, so the project I’m working on right now is building out an online class — ‘How to Start Your Own Country.’ In this class, because there are all these different approaches, people will pick their path: seasteading, a network state, a charter city, a micro-nation, or a sovereign state — which I don’t think we’re there yet, but if somebody wants to spec it out, I’m more than happy to help them. My personal attention is almost 100% on the charter city path. But they’re all different, there are things I find appealing about them. I may add startup villages — I kind of haven’t decided whether it’s too far off from ‘your own country,’ but maybe we develop a class for that later, because like you said, it’s so much easier and there will be so many more people doing it.

In the class, it’s teaching the history of this movement — there are just so many amazing stories of epic things that have happened: boats sinking in storms, pirates, the government of Italy blowing up the Rose Island platform. There’s all this amazing history. And case studies, learning the key mental models and frameworks — the importance of governance for GDP growth, what the founding team for one of these needs to look like, what skills you need, what your milestones are, how you should sequence things.

And then there’s your path — what type of project is it — and then there’s your ministry. The students, calling themselves founders, are going to roleplay as a government minister of one of the departments. Specking out a whole new society is way too much for a twelve-week class, so you’ll kind of pick which part of the government you’re most excited about, and in the later part of the class you’ll be developing a work product for that. Ministry of Commerce is build the economic engine. Ministry of Justice is what are the laws, how does the governance work — you’ll produce legal briefs, maybe write a constitution, dig into a particular area of law. One of my favorites is the Ministry of Story, which I’m using instead of marketing or PR because those are kind of superficial. The idea is: what are your founding myths? What is the story of why you’re doing this? What is the story of your society? And then go out and tell it the way we do in 2026 — with social media, videos, grow your audience. And there’s a Ministry of Community, which for a startup village project especially is about finding a group of people who have shared values, who all want to do this new society thing together, finding them and finding ways of bonding and figuring out what you can do together.

28:45 ISABELLE CASTRO

It sounds really fun. Even as someone who often thinks about what my network state or startup village would look like — and I haven’t fleshed it out at all — but even for someone like me who’s just interested in the movement, this sounds like a really cool way to learn more and actually experience what it means, and maybe even find your place within an existing one and meet other people who are also interested. Is there a social element to it as well?

29:17 PATRI FRIEDMAN

Yeah. There will be. We’ll have our own chat rooms and you’ll be able to meet other people in your cohort and work with them on things. I really think of it as teaching people, but then by the time that first cohort is done, now suddenly you have this community of engaged people who are all at a similar stage. One thing I’m loving about it is I feel like I lost a lot of the fun in the seasteading to charter cities transition. Those are the two halves of my career. Seasteading was just such an epic meme factory — I have half a bookshelf of books people have written inspired by seasteading. I run into art projects, it’s had an amazing cultural impact, I see it in TV shows. I switched my focus because I think the ocean is just a really difficult, harsh environment. That was the only option in the 2000s, so I did my best to figure out how it could work. But now that we can do this on land — where you’re closer to existing infrastructure, existing people, there are airports, all this stuff — I’ve been really focused on this, and it’s my chosen path towards trying to eventually build sovereign states. I have no regrets about charter cities as the path. I think it’s right. But there are not pirates, there’s not Kraken. A charter city is very explicitly under the host country’s constitution and sovereignty, so it lost a lot of the romance. I was kind of just okay with it because I want this to actually happen — but now that I’m in this class where I’m teaching everyone who’s interested in any of these approaches, I get to have the fun of these crazy stories and adventures that people have had while teaching things that I think will help everyone, whatever their approach.

31:41 ISABELLE CASTRO

No, I can completely see where you’re coming from — and it might even create its own meme in itself. Even as a writer going through that and going through the idea of the ministries, I can see that sparking a whole wave of inspiration. And was the ministries — was that AI’s idea?

31:54 PATRI FRIEDMAN

That was the ministries — that was AI’s idea. It’s the best idea I’ve gotten from AI in my few weeks of using it. I came up with the goal — I was like, I want people to focus on different work products. Some people are really going to want to write an industry analysis and some people are going to want to write a national anthem. I want to support them all. And the AI came back like: here’s what you should do — they should occupy the position of a government minister. Divide out the different things that society needs to do into ministries and teach them. I was like, ‘Wow, I love that. That is tight.’

32:36 ISABELLE CASTRO

Is there an element of you trying to find founders and kind of people to advise, to take on within their own ministries?

32:48 PATRI FRIEDMAN

It’s not the primary focus — it’s about reaching a mass audience and educating people, growing a brand using these modern tools of AI and short-form video, growing a brand around the broader space that I’ve wanted to do for a while, and forming this community. But absolutely — people whose focus and type of society best matches the type of things that I fund, of course I’m interested. And actually starting a company or doing anything is super hard. Most of the people who take this class are probably not going to keep on after the class and grow it more, but some of them will. And I’ll get to see their progress over time and have this community of engaged founders who I can bring to events in the space.

33:59 ISABELLE CASTRO

In a few of the talks that I’ve seen of yours recently, you’ve talked about founders in the space and how, with more experimentation and more negotiation with governments, some of these barriers to entry are actually receding, and soon founders might be kind of the next — I don’t want to use ‘hurdle’ again — bottleneck.

34:20 PATRI FRIEDMAN

Yes — bottleneck, they have been for a while. In the 2010s, if a government was willing to talk to me, I was super excited. In the 2020s, when people offer to intro me to governments, I mostly don’t follow up because my bottleneck has been founders — people who are willing to take on this epic challenge and have the skills needed to have a chance at doing it. My focus has been within this innovative governance world community, putting the word out about founders and asking people and trying to figure out who is intrigued but needs to be sold in order to become a founder. Bradford worked with me at Pronomos for like three or four years learning the space and advising our companies before he was like, ‘Okay, this is legit. This is real. And this opportunity in Africa is good. I’m going to do it.’ When I raise Pronomos Fund 2, probably next year, that’s going to be one of the focuses: who has the chops but isn’t sold on it yet? How can I find those people and sell them? My other plan for Fund 2 was: can we figure out how to identify new or young founders with high potential and put them together to learn from each other and see if the group brain — with people with less experience — can produce something? And that’s what this class feeds into — it’s casting a wide net and just letting anybody try, but by the end of it, you’ve created a work product. Maybe it’s an investor pitch or a description of the culture of your society. There’s something to evaluate them on, and see who’s making great stuff, and include them in future founder communities.

36:42 ISABELLE CASTRO

What goes into you identifying that they have potential? Like what made you think that Bradford, for example, was the one to do Alpha City?

37:00 PATRI FRIEDMAN

The track record is just the most reliable indicator. You’ve been in the arena, you slayed a pack of wolves, you did it again and took out a rabid lion or whatever. I think the much harder problem is for people who are earlier in their career — how do you figure it out? There are things that are different about founding in this space: having to deal with governments as your key partner, dealing with real estate and urban infrastructure. But I really think that there’s a huge overlap — if you’re good at starting a business in all the same ways that Y Combinator and VCs look for — not like coding, but sales and marketing and incrementalism and experimenting and focusing on the customer and grit and hustle — all these things are needed in our space. The biggest difference between a great founder in tech and a great founder in this space is: do they want this? Is this something that inspires them, that they’re passionate about, such that they want to try to do this newer, harder thing?

38:33 ISABELLE CASTRO

I guess it’s a lot more of a long-term thing. It’s more kind of like a dedication of life — which I’ve seen startup founders do, they do dedicate their whole life to startups, but they often just do one for like five years or so. How much do you think that longer time span plays into it?

38:52 PATRI FRIEDMAN

Ideally the founder is in it for decades. And yeah, that time span is longer. There’s a decent set of Silicon Valley founders who just stayed as their company succeeded for quite a while. I really think that your years of getting from nothing to having built something that is operating, profitable, and if it needs a government deal, it has a government deal — that is the hard part. If somebody spends five years doing that and has a thriving community with a growing economy and they’re like ‘okay, I’ve done the startup part, now I want to pass this on to a professional manager and step back’ — I think that’s okay. It’s this start that’s so hard.

40:03 ISABELLE CASTRO

But people are making headway, so there is optimism to be had, I think.

40:09 PATRI FRIEDMAN

Yeah. It’s really hard, but every year there’s more people, more companies, more news articles, more governments that have heard about this and are interested. We’re definitely getting there.

40:28 ISABELLE CASTRO

So going back to the class — do you have an idea of when it’s going to be finished, when you’re going to launch?

40:41 PATRI FRIEDMAN

It’s going so fast with AI and it’s so fun, and it just fits so well into my portfolio of what I’m doing. I’m aiming to launch this quarter — so by end of June. By the end of this month I’ll have the first few modules: the introduction, what are these paths and ministries, the key mental models, how this is going to work, an introduction to the space — and then at least one of the content modules to test out the whole system. Aiming for end of the month for that first test, and then launch sometime in June.

41:31 ISABELLE CASTRO

Okay, cool.

41:32 PATRI FRIEDMAN

Yeah, I’m trying to soft launch next week — just a website, social media placeholders, and a mailing list.

41:38 ISABELLE CASTRO

Okay, that’s going to be amazing. And I’d really like to test it out if I can. Thank you so much for coming on the podcast. This has been a long time in the making — I approached you like a few months ago to try and do this episode, so we finally got there, and I’m glad that it’s at such an interesting point in time, when you’re creating this course. Thank you. And thank you for watching, subscribe to the newsletter.

42:10 PATRI FRIEDMAN

Shes great.

Over the past few weeks I’ve noticed many of those around me have started to focus more on the analog life. Not that we were sitting around on video games before, but I’ve noticed an active preference to engage in activities together that take away from the machine and the screen. Music concerts with very few photos, card games and potlucks have become the standard state of play, phones remaining firmly in bags and pockets.

Beyond that, events and dinners that centre on network state building have focused on human connection rather than supercharging technological innovation. A dinner I attended at a ranch a week ago pitched their project as a “pro human” village, centred on reconnection with food production, the arts and human interaction.

Perhaps it’s just exhaustion. Many of my friends and colleagues live screen based worklives in tech. They are so plugged into the rapidly increasing pace of technology innovation in their day-to-day that discussing anything from that world in a social setting may just be too much.

So perhaps it was no surprise that I ended up, a few days after jigsaw night, in East Austin at a workshop titled “Intentional attention”.

The workshop

The workshop was tucked into a cosy upscale lounge filled with cushions and plants. There was a poker game going on in the room next door. Within minutes of arriving our phones were locked in the “phone jail” that lay on a coffee table in the centre of the room.

I had interviewed Dean Berlinerblau, the facilitator of the workshop, a few days prior. Coming to the interview expecting it to centre on a discussion on the regulatory landscape for psychedelic therapies, it soon turned to tech culture, AI and human attention spans. The attention economy, in his view, “so often dissociates us from what is deep and true and beautiful and disconnects who we think we are from who we actually are.”

The essay he’d prepared for the workshop opened with figures - 6 billion internet users, 5.7 billion social media users, 5.6 billion smartphone users, 1.2 billion generative AI users and counting. These technologies, he wrote, had become the default spaces “through which we socially relate,” and the attention economy had turned them into a mechanism for running “its ever-accelerating race to the bottom of our brain stems.” He described humans’ almost unconscious dependency to the tech, driven by the dopamine hits social media algorithms are meant to deliver. “We allow for it. We even enjoy it. And even when we don’t allow for it and don’t enjoy it, we do it anyway. Because we’re aware of how easily our resistance to the machine can be absorbed by the machine itself.” Creating the conditions to do nothing, he argued, had become a radical act.

Now, he stood in the middle of the room, explaining that the next two hours were to be focused on strategies to reclaim attention. “There has been no other moment in time where our nervous systems have been hooked so specifically,” said Dean. It shows up in mental health systems, exceeding just two hours of daily screen time on weekdays doubles the odds of clinically elevated anxiety, and quadruples the odds of emotional and behavioural difficulties. Half of American teenagers now spend four or more hours a day on screens outside of schoolwork, and among that group, roughly 1 in 4 reported symptoms of anxiety or depression, more than double the rate of those with lower screen time. “It’s a problem that is so urgent [...] that conviction to change can fly away in a matter of minutes, because these systems are so powerful in how easily they can hijack our attention.”

Share

We went round the room for introductions, stating our name and our relationship to digital technology. It started to feel like an addiction support group, despite most of us describing technology relationships that were pretty average today. “Hi I’m Isabelle” I said, “I’m on the internet all the time for work, I have to be, the news cycle is so fast now I might miss something if I don’t connect for too long.” Another participant described their morning doom scrolling habit, while another said “I’ve gone to reach for my phone twice already” (despite it being in phone jail). Then we went through three rounds of writing about how we would treat relationships and tasks differently if we didn’t have access to our phones. Each round was followed by discussion, which only served to clarify that that the attention economy was having an impact.

This isn’t, Dean was careful to say, simply about technology, the problem predates the smartphone by several hundred years. It’s a problem with “subject-object metaphysics,” he said, the assumption baked into Western thinking that inner life and outer life are separate processes and that optimising for the external life, in our work, gossip, our physical appearance and material possessions is not only fine but admirable.

In the interview, I’d raised Marc Andreessen’s well-publicised admission that he has essentially no inner life, no interest in introspection. Dean hadn’t been diplomatic about it, “That’s wrong. It’s not true or right that we should forego introspection. No one is exempt from mastering inner space in order to master outer space.” In his eyes, the introspection was a key component in the creation of systems that are positive for humanity, taking a break from the “run fast and break things” state of play.

The specific failure, as he saw it, ran deeper than Silicon Valley’s cultural preferences, now seeping into the design of algorithms to commandeer attention spans. What the attention economy had done to our nervous systems was “inevitable” given these foundations.

The workshop commitments sheet, which we got to near the end of the two hours and left space to write down the actions we would take to take a break from the attention grasping cycle, had framed it as a personal choice - “what you commit to,” it read, “should reflect who you told us you are at your best, not who the Machine has conditioned you to be.” In the room that felt clear. Outside it, with the news cycle running and the Telegram channels pinging and the agents messaging for permissions, it was harder to focus on.

More time or 996?

Dean had described AI as one of several “wild cards being deployed into society currently” which are being rolled out at scale without introspection around how they might impact society being properly addressed.

Leaders of Big AI have been telling us that AGI is almost here, AI agents will be in the workforce any day now and all of this will eventually give us more time for more introspection and “pro-human” pursuits. The claims escalate regardless of whether the previous ones actually materialise.

Although AI is progressing, these wild claims make use of the attention economy mechanism to drive hype and sales. An extreme claim generates coverage, coverage generates investment and more pivots and announcements, and by the time the claim fails to materialise, the next one is already in the cycle. The prediction doesn’t need to be true, it just needs to make waves.

Share

The hype has a way of eating into everything, even those places you would least expect. Allbirds, a footwear company, recently announced a pivot to AI infrastructure under the name NewBird AI, not because it makes obvious business sense for a shoe company, but because AI is where the attention is. The announcement caused shares to rise by 600% – a complete turnaround for a company whose stock price has slowly dwindled. Many similar announcements have been followed by layoffs, adding a layer of real world anxiety to what is otherwise largely driven by online discourse – people’s working lives disrupted not by what AI can actually do yet, but by what companies believe they need to be seen doing.

Aside from this, the AIs themselves are often built to sustain attention. The constant follow up questions and suggestions, along with positive affirmations, create their own cycles of dopamine and engagement.

Founders in the AGI space have long assured me that AI’s increasing capabilities would give us more time away from the computer. But employers increasingly voice preference for workers who embrace AI workflows and “996” (9am-9pm, 6 days a week) is becoming the dominant work culture in Silicon Valley, not the 4 day work week. Those I know who have embraced AI’s capabilities seem to have less and less time away from their computers, filling the time they’ve “saved” with more agents, more output and bug fixes, keen to demonstrate the efficiencies of the algorithmic workforce. The promised clearing of time never quite arrives, unless you count mass unemployment, which is a far cry from the utopian ideal AGI leaders depict.

Counterbalances

But if humans are anything, they are adaptable, and within the AI race madness, a counterculture may be emerging driven in response to all the noise.

A friend of mine just moved to SF in pursuit of funding for her new project – a live-in accelerator program. As I do when arriving in a new city, she had joined a coliving house, filled with developers and tech entrepreneurs. She had sat down for dinner one day, joined by a developer at a leading design software company. Distressed, the developer voiced her concerns about Anthropic’s latest model release, which includes a design tool – if good enough (which it might be, Anthropic previously had two seats on the design company’s board) it had the potential to seriously undercut their business. In the days since the model release, the stock of both Figma and Adobe has dropped like stones.

The dinner coincided with a discussion my friend had had with an investor around the same time who had noted his declining excitement with new software. “Anthropic will eat it all,” he had told her. His investment thesis had therefore shifted to focus on more human-centred businesses, focused on fitness and well being. To counteract the AI slop, he reasoned, the average person will have to focus more on what they can do that AI can’t.

I left the workshop and checked my phone, of course. Aside from getting an Uber, I had people to message and a Duolingo streak to maintain. The dinner guests at the ranch are still building their network state, which will have an app and potentially a cryptocurrency. Even Dean was taking photos within the workshop where we had no phones to potentially use for future promotion.

This isn’t a story about people rejecting technology completely. Most of the people at that workshop, at that dinner and around that jigsaw table are in the technological front lines. They work in tech, use AI, and, in the case of network state founders, are looking for ways it can be integrated to “improve” existing systems. “The Machine” is the water we swim in now, to reject it completely avoids the benefits it can and does bring.

But the kind of intentional attention that Dean was trying to cultivate in that East Austin lounge puts a stop valve around passive consumption. It makes us more aware of when and where and how much. In an economy so often built on harvesting that awareness, that’s the best leverage any of us actually have.

Share

Food for Thought

• Young people have been found to be ditching smartphones for dumb phones, joining tech-free clubs, and booking digital detox stays in growing numbers. Fortune tracks how the backlash against screen culture has become its own market with Offline Club now in 19 cities and searches for analog experiences doubling in 2025.

• Silicon Valley AI startups are openly requiring 72-hour, six-day work weeks from employees, interestingly, the same schedule China’s Supreme Court ruled illegal after worker deaths. Fortune wrote about the economic pressures driving the trend and the researchers already warning about its sustainability.

• “Marc Andreessen Is Wrong About Introspection” by Joan Westenberg, places Andreessen comments within intellectual history, arguing that what Silicon Valley calls a productivity hack is actually the abandonment of the thing that makes good judgment possible.
  
  

  FAQ

  Q: Is AI actually giving people more free time? A: Despite promises from AGI leaders that AI would free up human time for richer pursuits, the evidence so far points the other way. Workers who adopt AI workflows tend to fill saved time with more output, more agents, and more bug fixes. Meanwhile, 996 culture — 9am to 9pm, six days a week — is reportedly becoming the dominant pattern in Silicon Valley, not the four-day work week.

  Q: What is the 996 work culture and is it spreading in the US? A: 996 refers to working 9am to 9pm, six days a week — a schedule China’s Supreme Court ruled illegal after worker deaths. It is now reportedly becoming the dominant work culture in Silicon Valley AI startups, driven by economic pressure and the expectation that AI workflows should produce more output, faster.

  Q: What is “intentional attention” and why is it gaining traction in tech? A: Intentional attention is the practice of consciously deciding when and how to engage with digital technology, rather than defaulting to passive consumption. It’s gaining traction among tech workers partly as a reaction to the speed of AI development, and partly because research increasingly links high screen time to anxiety and emotional difficulties — exceeding two hours of daily screen time on weekdays doubles the odds of clinically elevated anxiety.

  Q: Is there a counterculture emerging against the attention economy? A: A counterculture is quietly forming inside the same tech circles that built the attention economy. Analog socialising, phone-free dinners, digital detox events, and human-centred investment theses are becoming more common. Some investors are specifically shifting focus away from software toward fitness and wellbeing businesses, reasoning that as AI generates more digital noise, humans will pay a premium for things only humans can do.

  Q: How does the attention economy affect mental health? A: Research cited in this essay shows that exceeding two hours of daily screen time on weekdays doubles the odds of clinically elevated anxiety and quadruples the odds of emotional and behavioural difficulties. Among American teenagers spending four or more hours a day on screens outside of school, roughly one in four reported symptoms of anxiety or depression — more than double the rate of those with lower screen time.

But what happens when you apply the prediction market logic to how organisations actually make decisions? That’s the question at the heart of UMIA.

I was initially sceptical. When prediction markets first gained wider attention, they felt adjacent to the memecoin casino end of crypto - a sophisticated-sounding wrapper around speculation. What I found in speaking with Francesco Mosterts was something more substantive: a framework that attempts to repurpose market logic to make governance more rational, more resistant to capture, and more accessible to the broader market rather than a handful of large token holders.

We covered decision markets versus prediction markets, the problem with token voting, how off-chain legal structures enforce on-chain decisions and whether any of this can work for non-crypto organisations. If you’re curious about new models of capital formation and coordination, this one is worth your time.

Share

Topics covered:

• From prediction markets to decision markets what’s different and why it matters

• Why token voting is corruptible and how decision markets address that

• How the broader market brings rationality to governance

• Concrete example: an AI baby learning app raising capital via decision markets

• Who decides what goes to a decision market and what stays with management

• On-chain and off-chain enforceability, including the UMIA Governance SPC

• The three levels of safeguard against bad-faith decisions

• Capital formation for Web2, AI-native, and public goods projects

• Where prediction markets are headed and UMIA’s place in that future

Transcript

1:04 Isabelle Castro

Hello and welcome to the Utopia in Beta podcast and today I’m joined by Francesco from UMIA and we are going to talk about using prediction markets for governance which is a really interesting concept. Like this was pitched to me and I was like, wow, I’m just getting my head around prediction markets in general and how -- like I think like a lot of people when they first started coming out, it looked like another kind of form of the memecoin gambling kind of vibe. So I’m really keen to hear more from you about its utility and why it’s good to use this instead of other governance things. But to start with, maybe you give me an introduction to yourself and UMIA and how you got to where you are today.

1:56 Francesco Mosterts

Yeah, sure. Hi, Isabelle. Thank you for having me here on your podcast. It’s really great to start explaining to people how actually this prediction market can become a useful thing for humanity and for coordinations between parties. And yeah UMIA is really our idea around that and the way that we aim to solve a big problem in the crypto ecosystem. And the idea really came from the lab behind UMIA which is Chainbound -- the company where UMIA was incubated from -- that is a laboratory that focuses on crypto, on Ethereum in particular, and on infrastructure solutions that enable blockchains and in particular Ethereum to do more, to become more useful, to become more entrenched in everyone’s lives. And yeah that’s mainly the focus of it.

2:56 Isabelle Castro

Okay cool. And so -- I was when I was pitched it I was told that it was using prediction markets to run organisations. Take me through how that works. From step one basically.

3:15 Francesco Mosterts

So the big thing -- let’s start with the premise of how we go from prediction markets to decision markets and then I’ll tell you how that fits into as a tool inside this big machine which is UMIA. So prediction markets have the ability practically to price future outcomes instantly. So the outcome could be a binary outcome -- yes or no -- and you price the yes or the no. That could be anything from like is X going to be the next president of the United States. You put a price to it. You sort of go through the common knowledge of people, what’s the probability of that, and put a price behind it because people put money. So that’s a great coordination mechanism at a high level.

Subscribe now

And what we did with that -- we said okay but you can price whatever prediction market wants but most importantly you can price decisions and you can price the impact of a single decision and that’s very very unique. And a case is you are a company and you have governance and you want to make a decision out of it. You can make a token voting out of that which is literally whoever has a lot of votes says yes or no and they easily manipulate the direction of that token in the strategy that they like. Or you can use something which is very similar to prediction markets which is the decision markets in which you say okay we want to make a decision and people can price the outcome of these decisions and people can trade and speculate on these outcomes. And an example is you have a token trading at 10 and you have a decision market that says should the token do X or Y and X could be issue extra 100 tokens to Paradigm because they are a great VC, or they should not. That’s a simple yes or no. And then people trade on this. And anyone from a sophisticated to a less sophisticated investor could say well fundamentally the token price with this should be 110 so the yes plus token should be 110. The no should be 80 because it means that someone else would take that money and they will lose market share and so on. So you have a very very simple way for people to price the future outcome of decisions and allow on one side everyone to be incentivised and interested to express their preferences for a given governance proposal -- hence make money -- and on the other side you have the team that has a very clear and less corruptible way to decide the future of a proposal.

So that’s a very very high level how it works. And how does it fit into the UMIA protocol -- UMIA is this system that allows projects to launch a token, so we’re going to have our own sort of launchpad, and then have that token fully governed by decision markets. And when I say governed I mean you need them for stability because otherwise -- the example I made before -- the founder could just say well you know what, the market said I had to issue 100 tokens more because the yes is higher than the no but I’m not going to do it because I think it’s better not to do it. You need enforceability. So onchain enforceability is easy because it’s just code. So you have an easy way to say if the yes is high, automatically this function onchain will be called and so that happens. And offchain we need an offchain enforceability and that happens with the legal framework that we are building in partnership with Metal X which is one of the best legal firms on this setup of things.

7:07 Isabelle Castro

No that’s really really interesting. Okay I’ve got so many questions around that. First of all is kind of the relation to price and how do you avoid those with the most money shaping the decision because they have the most money to play with?

7:33 Francesco Mosterts

That is first of all a very unique part of decision markets which is -- if you are a large token holder you have a disproportionate amount of control over that single token and that single project. You know, you can take the other proposals -- the main team has majority of the shares of the token and they will direct it wherever they want. But here the beauty of decision markets is that anyone can trade in the different conditional markets. It’s not just people that own the token, but even a trader, a highly sophisticated hedge fund or even a not very sophisticated person, a discretionary hedge fund that just has dollars and says if this decision ends in a yes -- so if this conditional market lands on yes and is actually settled -- I want an exposure in this token. So you have people buying tokens conditional on future decisions, hence making token price a decision and a new asset class.

And the beauty is now you have not just token holders -- so a very small share of people managing it -- but you have the broader market, the broader rationality of the market pricing this and giving a fundamental price to the two things. So again same example I’m making again and again -- fundamentally it trades at 110. If the thing starts to trade at 130, people that have fundamentals would just say, well, this is overbought, I’m going to sell it and bring it back to 110. So if there is rationality, then you have that extra in the mix to drive the actual value. So the economic aspect is very very powerful here into bringing rationality and non-corruption from large token holders into what they want. Still there is a risk.

9:27 Isabelle Castro

Okay. So does the economic thing -- do you believe that it brings kind of incentive to act in the best interest?

9:36 Francesco Mosterts

100%. Because first of all there is a direct incentive to act in the best scenario for the project because whatever -- if you are a token holder you would trade in the market that is a conditional market that then settles and if it settles at a higher price than the one you bought well you made a sort of arbitrage, a profit over your position that was conditioned and it settles. And on the other side, it doesn’t really make a lot of sense to trade a market in which you, for example, buy a conditional market that says you should do X and that X is actually bad for the token because you are spiking up the price of the token knowing that as soon as it’s settled, it’s going to drop to zero -- to nothing. So you are making exposure in a token that on top is going in a direction that is let’s say stupid and so you are doubling your chances of losing money. So again the rationality, the economic incentive and the simplicity of the setup could drive a lot of good decisions.

10:52 Isabelle Castro

Okay. And do you see this kind of governance structure being applicable to anything beyond kind of token governance, for example like more real world organisations?

11:08 Francesco Mosterts

100%. The beauty of all of this -- what is the division between what’s the tool and what’s the product. The tool of UMIA are the tokens, the launchpad, the decision markets. Those are tools to get to a product and the product is a new framework of capital formation and coordination. Capital formation through the token setup that we have and coordination with decision markets. That’s the product. Does it apply to web two? Does it apply to other things? 100%. And there is a very interesting chain of thought and research arm inside of UMIA that we are exploring which is in an AI first world in which we are going to have a lot of solo founders or small teams -- very small teams -- having ideas and real applications of LLMs, for example specific agent use cases or specific hardware with certain things and so on. Well, they cannot -- the speed at which venture capital goes and the speed at which normal capital allocation goes doesn’t really fit that because they could literally just want a way to raise money, have people have shared ownership of whatever they are building, and on top of that have a sort of recurring expense setup which is the opex on LLM tokens. And so with that sort of setup it fits really really well this capital formation framework that UMIA is. So to answer your question 100% -- the tokens, the web three setup -- the way that we believe it is a tool is not the end goal. The end goal here is capital formation and capital formation can happen for web three companies, on chain companies, AI agentic specific companies, web two, public goods, collectibles, and anything that benefits greatly from shared ownership and decision markets.

13:17 Isabelle Castro

Okay. Would you be able to take me through kind of a scenario in a web two company or kind of like a public goods situation where this could be implemented and how it would work, just for some tangibility basically.

13:36 Francesco Mosterts

Yeah. Let’s make an example of an entrepreneur from -- let’s take Italy, I’m Italian -- from Italy. He has a great idea in which he wants to build an AI agent for people to -- you know -- have a different tool for babies to play with. And this is like a game that helps babies learn how to speak. Let’s do this. Okay. So they don’t have connection to VCs. But the idea makes sense and the system has an initial sort of cost to set it up and then a very recurring cost of opex which is the expense on LLM tokens and the subscriptions and so on that goes to OpenAI and similar. And he says well I really need first of all the first 100 users but on top of that I need a way to raise money and I need to do it fast because I know how to do this. This doesn’t take a team of 10 and so many people. I just need good distribution marketing and I know how to do it. So I just need money to cover the first two.

Okay. So they set up -- we called it a quantum token -- which he raises, let’s say, half a million dollars. And with that he says a part of it is going to be used initially for the distribution, so he has direct access to a part of it and he can set up in the framework that we have. And then he needs a monthly disbursement of 50k a month to cover the opex cost of LLM requests. Right. So with that he sets it up, he has the token, he does this token fundraise and people that believe in this sort of things buy the token and they have direct exposure in the future potential success of this company. He sets up the legal structure through the UMIA SPC which is the legal umbrella that we built in which fully delegates ownership to decision markets. And then he has this monthly disbursement that goes directly to OpenAI -- and there’s an invoicing mechanism that then happens with a technique called CKTLS but we don’t go into details -- but anyway there is a way to be sure that those money got used for what he said they’d be used for. And then yeah he just builds the company, the legal framework ensures that token holders actually accrue value and it’s not just oh I have a token that somehow is connected to this toy company. No -- the token holders request sort of informations and so on and everything gets accrued to the token value. It could be through buybacks, it could be through a vault mechanism and whatever.

But with that he would say well I raised money in 5 days or 7 days because my idea really made sense. I have my first 100 users which are also the token holders. So they have upside. So they bought into me instead of, I don’t know, the Lego or whatever else there was in the market because they were like, I own something and I also now have a game for my kids. And on top of that, he has an easy way to pay the initial bills and then if things go great and this becomes a million dollar business, he can migrate to another setup and anything that would make more sense for him. So yeah, that’s the case in which I would say we really solved the problem. Was crypto the focus? No, the crypto was the tool and that’s what I really really stress about.

17:18 Isabelle Castro

Okay. So it’s kind of like crowdfunding with decision markets.

17:27 Francesco Mosterts

Yeah.

17:27 Isabelle Castro

Okay. Interesting. With the decisions that go into the decision markets, who chooses what is to be decided on? I only ask because in kind of other distributed mechanisms of governance like DAOs, too much decision-making can get really complicated and slow down decisions. How do you avoid that and move away from those kind of issues?

18:04 Francesco Mosterts

100% I agree that the centralised decision making on the day-to-day is a mess and is also not incentive compatible. Decision markets are there to replace board level decisions. Nothing else. The CEO and the management and the operators they run the business as they would run any business. The only difference here is the treasury of the business is in the hands of decision markets -- it’s not in the hands of the actual founders. So you have this sort of difference in trust and accountability which is good for -- you know the example I made before of the Italian founder -- no one really knows who he is but you have less trust to put into him because he doesn’t really have access to all the money, he has access to this disbursement that goes to him. So that’s the big first thing.

Share

And the second thing is we need to give customisability for the set of businesses that come on board on how the decisions get created, how the proposals get created and how they get decided. So there are sort of companies in which it would be okay to have a fully permissionless way to set up decision markets -- and anyone can set it up -- and that’s probably a more web three native company, they’ll be okay with that type of thing, that are okay to open up the decision making process to a maximum level. And then anyone can really trade -- and it’s impossible to put any constraint on who can trade because it’s the big piece of decision markets. Then there are more traditional companies in which we would either have a minimum stake of the native token to get in, or even just allow the management to open the decision market.

And the decision markets direct a few very specific decisions which is management of treasury -- so increasing the disbursement, doing investment from treasury -- and then changing token supply, minting or burning, changing management, and changing strategy. So anything really which is at the board level -- a board would sit around any of these decisions. Anything else is the team that has to do it.

20:48 Isabelle Castro

Okay. And I’m keen to hear more about the enforceability as well -- especially offchain -- because I feel like onchain that can be enforced through code and smart contracts and all those kind of things. Offchain, you said that you’re working on a kind of legal structure. How does that work? And how can you make sure that it does work in the best interest of the company?

21:21 Francesco Mosterts

Yeah and that’s the beauty of -- I think one of the beauties of UMIA -- that we are innovating also there in the legal aspect because UMIA has this UMIA Governance SPC which is an overarching entity which you could see as like the big bucket that then holds all the projects that launch on UMIA. And this big sort of bucket has a few rules for the projects that launch. One is that they are required to delegate governance -- so board governance -- to decision markets.

Then any project that launches on UMIA, they launch on top of the SPC. They are sub-companies -- that’s what they’re called. They have a segregated balance sheet and full legal accountability -- meaning they are at all effects a company -- and they have zero accountability or liabilities related to the rest of the companies. But the only constraint they have is what I said -- the decision market. And that’s really the fact that allows us to ensure that projects launching on UMIA -- whenever a decision market doesn’t have enforceability onchain but is offchain -- well, we know that it’s going to be enacted. Otherwise there is a very easy proceeding for token holders to say this was decided onchain, this was not done by the team, we require it to be followed as if it was a board decision.

So that’s high level how it works. And the cool thing -- the innovation where I said the innovation comes -- is that this is actually a setup that’s used by mutual funds. The SPCs normally are used for mutual funds. You have the big company and then you have the single funds that are single sub-companies and instead here we use it for something else. We use it for projects that have different governance. So yeah, it’s going to be very interesting to see how teams play around with it and how the community to project relationship builds in terms of any offchain decisions that could be contentious in any way.

23:38 Isabelle Castro

Yeah. Because Utopia in Beta looks at kind of the governance experiments that are happening onchain and online and then looks at how they can be in the real world and in the creation of new societies. And I’m thinking -- in the creation of new societies which are founder led -- I know at least a few founders that they would offset decisions but maybe wouldn’t they have a particular way that they want to take which is more of a long game that maybe the community can’t see. And I just wonder how that works around -- okay you put the decision to the community, to the decision markets, but you don’t actually like the way that the decision’s gone -- or you’ve just got a whole group of trolls -- which I realise that you’ve said that there are incentives to not troll and to work in the best interest of the project. But if for some reason you’ve got some really kind of well-funded trolls that really want to just mess with stuff, is there a kind of kill switch or mechanism for not following the decision market?

25:05 Francesco Mosterts

Yeah. So let’s start with three levels of safeguarding.

The first level is ethical and legal -- in a sense if a decision is heading into something that’s actually illegal there are clear safeguards to not go there and it could be funding something that should not be funded or giving money where it should not be given. That’s the first level and that’s no discussion on that.

Then there is a second level which is -- as you said -- something irrational for the team to go forward with. And that’s the market -- and we really truly believe that the market always drives the maximising welfare function to the token holders.

26:12 Isabelle Castro

So ethical, the market -- and those are the two main things. And then there’s another thing but -- yeah.

Okay cool. Well we’re coming to the end of our time -- I realise I only scheduled for half an hour -- but before you go I just want to hear a bit more about -- you said that you’re doing research and continuing development. Where are you positioning yourself going forward in the next couple of years and how do you think the rise of prediction markets might influence that?

26:46 Francesco Mosterts

Yeah. So where we want to position ourselves is to be the best sort of capital formation mechanism onchain that exists -- to have a solution for teams and projects that come from different backgrounds, have different types of ideas and requirements, to come on board and launch their project. And we believe that projects that are community first -- so projects that really require the first 100 users and live and breathe because of that -- would rather decide into this. And also we aim to prove that decision markets are a great governance mechanism for this specific niche. And yeah we think that is the biggest thing that we aim for -- to build a solution that allows tokens to be a welfare maximising sort of financial solution. So our vision is we move from a world in which people say well tokens really are most of the time memes, to a world where we say well tokens actually are a great programmatic solution to transfer value and to accrue value in a given project idea and so on. So that’s the high level vision.

And what we see with prediction markets -- prediction markets again are, in my opinion, a great pricing logic into very exotic stuff that could be from the probability of something and so on. And it’s a natural progression -- prediction markets started from just betting on you know who’s the next president and so on to now being more sophisticated. We see that then spanning across different types of industries and obviously governance is the one where we believe there’s a case. The connections there remain that we simply believe that human cultural interest is going to move towards having prediction markets as a governance framework -- in fact decision markets.

29:10 Isabelle Castro

Okay. I feel like I could go into so many different topics around this, particularly as you mentioned its increased importance around AI development and kind of solo entrepreneurs. So we’re going to have to do another call at some point, but thank you so much for coming and sharing what you’ve been working on and how it could impact governance going forward. It’s been super interesting.

29:33 Francesco Mosterts

Thank you. This was great.

29:40 Isabelle Castro

Okay. Thank you.

It starts (and ends) in the coastal region of Emilia Romagna, Italy, where, in 1958 engineer Giorgio Rosa decided to stake his claim on international waters. “My initial project was to build something that could be free from any constraint, and wouldn’t require a lot of money,” he reportedly noted. “On dry land, bureaucracy had become suffocating.”

Subscribe now

Over nine years, Rosa funded, designed, and built a 400 square metre platform 6.27 nautical miles off the coast of Italy, an area technically deemed “international waters”. Completing construction in 1967, he topped the artificial island with a bar, a post office and a shop. A testament to freedom, the island lived outside the taxation of Italy, carrying its own flag and currency, even implementing Esperanto (an international auxiliary language created in 1887 for cross cultural communication) as the local language. Rosa declared sovereignty on May 1 1968 - a month imbued by anti-capitalist and anti-authoritarian protests in France - and pronounced himself the president of the aspiring nation.

“My fondest memory is that of the first night, on the island under construction,” he recalled. “Along came a storm, and it looked like it would tear everything apart. But in the morning the sun was shining, everything seemed beautiful and possible. Then trouble began.”

The trouble came in the form of the Italian Navy. The Italian government, concerned that Rosa Island was acting as a ploy to avoid taxation while raising money from tourists, sent the national police force to occupy the island and set up a naval blockade. On 26 June 1968, all residents were evicted from the island and the structure was demolished in February 1969.

Shortly after the eviction a new set of Rosa Island stamps were created, complete with a declaration (loosely translated from latin)- “Enemy violence destroyed the work not the idea.”

The short life of Rosa Island went largely forgotten for a few decades after its destruction - confined to the niche lore of those that longed for freedom and dreamed of making a country of their own. However, in 2020, Netflix released a film based on the story, which brought Georgio Rosa’s dream back into centre field.

Between 1969 and 2020, there have been multiple attempts at achieving sovereignty and the building of new, autonomous states. During those five decades, the Honduran ZEDEs were born, as was the concept of Seasteading that lent into the idea of constructing new nations in international waters, along with various other shots at nationhood using unclaimed land (Liberland being one of the most famous).

But the reason why I bring up all of this, is due to one part of the storyline in Netflix’s retelling of Rose Island (and I reference the film specifically because I’m unable to corroborate from historical sources that this is what happened in real life). In the film version, we follow Rosa to The Council of Europe. Fulfilling the three of the four requisites for statehood – a population, a defined territory and a government – he goes in search for the fourth (and most elusive) diplomatic recognition. Ultimately, he is denied.

While this is a fictional depiction of the story, and we now all know the island was, indeed, destroyed in real life, it points to a thorn in the side of statehood - formal recognition and the capacity to enter into relations with other nation states.

The myth of “Terra Nullis”

This brings us to the Montevideo Convention. Signed in 1933 at the Seventh International Conference of American States in Uruguay, it is the closest thing international law has to a checklist for statehood.

Article 1 sets out four criteria that a territory must meet to qualify as a sovereign state: a permanent population, a defined territory, a functioning government, and the capacity to enter into relations with other states. The first three are pretty achievable through will and resources, Rosa found that. The fourth is different. It means that the nation only really exists in its own right as a new state if other nations agree that it exists. Fulfil criteria one to three without it, and you are what international lawyers call a declarative state. A private club with a flag. This is the wall Rosa hit. It is the wall every network state or the like really hits when trying to achieve full autonomy.

Over the past two months I have been making my way through The Start Up States, a book written by real estate developer and early Bitcoin investor, Andrew J Starr. It’s a hefty volume, (although on its website it has a handy video and podcast to take you through the main points) which outlines a strategy for true sovereignty in the eyes of international law.

Before getting into Andrew’s proposed solution, it’s worth understanding why he thinks everything that has come before it has failed and his assessment is not especially generous.

The romantic vision shared by most would-be nation builders is some version of the same thing - find a gap in the map (or our perception of statehood), plant a flag and dare the world to argue. The problem, Andrew argues, is that the gap doesn’t exist. Every stretch of land on earth is either formally claimed by a nation state or in ongoing border negotiations between existing powers. The frequently cited case of Bir Tawil, a small, barren strip of territory between Egypt and Sudan that neither country formally claims, owing to a quirk in competing colonial-era treaties, is a favourite of aspiring sovereigns on the internet (did anyone hear that story about the guy who wanted to claim it to make his daughter a princess?). However, the absence of a claim is not the same as an open invitation. Egypt and Sudan have each made a decision, for their own diplomatic reasons, not to assert ownership as it left them open to declaring a more valuable piece of land as not their own.

The sea, in Andrew’s eyes fares no better. Vessels operating on international waters are required under international law to register under the flag of a recognised sovereign nation. Without one, you are not a floating republic, you are, in legal terms, a pirate ship open to all manners of disputes.

Digital states, for all their ingenuity, run into the same wall from a different direction - international law is fundamentally rooted in physical territory and the ability to enforce rules on land. If they can’t do that, Andrew’s blunt view is that they are not a “sovereign digital nation” they are an enthusiastic online community.

All of which leads to what he considers the more uncomfortable truth about many of the most celebrated micronation projects. They are, in his framing, LARPing – “Without recognition you are a private club with a flag.”

A new kind of sublet

The argument outlined in Startup States goes beyond the attempts at claiming Terra Nullis (unclaimed land), international waters or even the hallowed Special Economic Zone. An SEZ is a fatal flaw, it claims, perpetually reliant on the host country’s government - why go for bronze when you could go for gold?

Andrew starts with positing the startup state as a concept not much different from the tech startups which are now staples in our phones:

“Uber is the largest taxi company but it owns no cars, AirBnB is the largest rental company but it owns no property. A startup state can be a country that doesn’t own the underlying land title.”

The premise leans on the fact that sovereignty and land title are two entirely separate legal concepts. They don’t have to come packaged together and historically, they often haven’t. In this theory, startup states could lease the land from the host country for 99 to 999 years, structured around an international treaty in exchange for complete autonomy. The new zone is recognized as sovereign using the mechanisms of international law – a structure often much more resistant to changing political administrations than the agreements that underlie special economic zones. This would mean it could have its own immigration rules, its own tax system, its own legal framework and its own foreign policy. “A sovereign sublet.”

Andrew makes parallels with the micronation of Monaco, positioning its autonomy to make attractive laws as the asset itself:

“Cross from Beausoleil into Monaco. The views are identical. The sun is identical. The coastline is identical. The property prices are not. Monaco is expensive because it is a sovereign principality. The legal status of the territory is the asset, and that is precisely what the treaty creates.”

So why would a host country agree to this? Here the book leans on the work of Peruvian economist Hernando de Soto, whose landmark study The Mystery of Capital argued that the developing world is sitting on trillions of dollars of what he called “dead capital.” Many nations hold enormous quantities of land but lack the capital, infrastructure or institutional capacity to unlock its value.

Share

Through granting autonomy, albeit through a lease, it allows for external investment and intentional design to develop the land to become a revenue generating asset. The host country receives a rent for the land and can negotiate shares, therefore benefitting from any of its production. A non-performing asset becomes a high-performing one without relinquishing the land itself.

Selling that land risks secession disputes and territorial loss. Leasing it does not, because ownership and title remain with the host state throughout. It’s the kind of proposed “win win” I looked at in my journey back into medieval Europe a couple of weeks back.

Founders’ choice

The legal standing of these kinds of projects is important, because then it shifts to the people that build them - the thing that creates the foundation for network states and the like to move beyond existing systems.

In a summit I went to recently, Patri Friedman noted that founders had become the main limitation he had found in his own support of charter cities - “Our biggest bottleneck is incredible founders. Startups are hard and this is even harder. It’s a startup where we need to be talking to heads of state and putting together not only a real estate project, but also an economy and finding the right tenants to make jobs.”

I’ll go into that summit and my takeaways from the discussions that happened around it in a later newsletter but Andrew rests on a similar conclusion - The world map has become, always was really, a snapshot of time rather than an end state. The limitation now is founder talent and drive.

This brings me back to Georgio Rosa, sitting on his artificial island that had survived the physical storm with the hope that “everything was beautiful and possible”. His sovereign dream was destroyed by the Italian Navy, but a future where that doesn’t happen could now be in sight. What “everything” means for these new states may soon be entirely what these founders (and hopefully their community) decide.

Q: Why do most new nations and micronations fail to achieve legal sovereignty? A: Most fail because of the fourth Montevideo Convention criterion (1933): the “capacity to enter into relations with other states.” This requires diplomatic recognition from existing nations — without it, even a territory with a permanent population, defined land, and functioning government is what international lawyers call a declarative state. As one framework for new nation-building puts it: “Without recognition you are a private club with a flag.” Rosa Island, Próspera, and digital network states have all hit this wall.

Q: What is the Montevideo Convention and what are the four criteria for statehood? A: The Montevideo Convention (1933) is the foundational checklist in international law for statehood. Article 1 sets four criteria: a permanent population, a defined territory, a functioning government, and the capacity to enter into relations with other states. The first three can be achieved through will and resources; the fourth requires diplomatic recognition from existing sovereign nations — and it remains the decisive obstacle for every aspiring new state.

Q: What is a startup state and how is it different from a special economic zone? A: A startup state, as proposed in Andrew Starr’s The Startup States, is a territory that leases land from a host country via international treaty rather than claiming or conquering it. Unlike a special economic zone or charter city — which remain legally subordinate to and revocable by the host government — a startup state would have its own immigration rules, tax system, legal framework, and foreign policy, protected by international treaty law. The core insight: sovereignty and land title are legally separate concepts. You don’t need to own the land to be sovereign on it.

Q: Why would a sovereign nation agree to lease its territory to a new autonomous state? A: Drawing on Peruvian economist Hernando de Soto’s concept of “dead capital,” the startup state argument holds that many nations own enormous tracts of land they cannot monetise. A sovereign sublet turns a non-performing asset into a revenue stream: the host state receives rent and negotiated equity without relinquishing land title — avoiding the secession disputes that outright sale would trigger. The new state gets treaty-based recognition; the host gets development without territorial loss.

Q: What was Rosa Island and why was it destroyed? A: Rosa Island was a 400-square-metre platform built over nine years by Italian engineer Giorgio Rosa, 6.27 nautical miles off the coast of Emilia-Romagna. Declaring sovereignty on May 1 1968, Rosa established a bar, post office, and shop, issued its own stamps and currency, and implemented Esperanto as the official language. The Italian government, concerned about tax evasion, sent the navy to occupy the island in June 1968. It was demolished in February 1969. A Netflix film in 2020 revived interest in the story.

Expecting conversations solely around governance and network state creation I went along. What I found was quite different. Yes those conversations were very much there, but so were talks on philosophy, art, wellness approaches that were decidedly “woo woo” and a whole kids club for families to get involved.

Six months later I met Timour Kosters, Edge’s co-founder at Network School where, over breakfast, we discussed the ethos of Edge and what was behind the all encompassing approach. It all came down to creating community culture, and places that people want to stay in for generations.

While the team has focused on pop-ups, they stand at the brink of creating more permanent communities. His learnings are feeding into that transition, so I thought I’d get him on the podcast to see how he sees the culture evolving.

For anyone interested in building or participating in new cities, or maybe even just want to make their existing community better within the landscape of increasing loneliness and disconnection - this one is for you

Share

Topics covered:

• Why network states and intentional communities matter now

• How multidisciplinarity shapes culture and what Edge does differently

• The four core principles guiding Edge City’s villages

• Why multigenerationality is a feature, not a complication

• Integrating local communities, lessons from Patagonia and beyond

• Pop-up villages as incubators for longer-term permanent hubs

• Accessibility, scholarships and the criticism of exclusivity

• The role Edge City plays in the wider network state ecosystem

Transcript

0:34 Isabelle Castro

Hello and welcome to the Utopia in Beta podcast. I’m Isabelle Castro, your host, and today I’m joined by Timour Kosters, co-founder of Edge City. How are you doing today?

1:09 Timour Kosters

I’m doing really well, thank you. Thanks so much for having me.

1:11 Isabelle Castro

I’m so excited to talk to you. I feel like Edge operates in such a unique position within the network state movement. I’m keen to go into some of the themes it touches on, but to begin with, maybe we start with a bit about you and your journey to network states and how you came to be where you are today.

1:35 Timour Kosters

Definitely. So just a very brief background — I have a math and economics academic background and spent about eight years in New York building startups. Then I was running a VC fund called Seed Club Ventures. While I was there, I was invited to an event called Zuzalu, which I think was an origination point for a lot of this, at least the current state of the movement.

Zuzalu was started by Vitalik Buterin, the founder of Ethereum. The idea was that at that time — late 2022, early 2023 — there were a lot of ideas circulating in this ideological milieu. The network state book had just come out, and some of those ideas had been around for decades, but Vitalik’s insight was that nothing was actually happening. There were no real experiments being run, and he decided to plant a flag and run one.

I got involved because Vitalik shared the idea with a few people, those people shared it with others, and then they picked around ten or twenty “inviters” who would bring their communities, keeping things somewhat diverse in terms of ideological areas. Someone got in touch with me and said, “Hey, this thing is happening — they’re looking for someone to run the AI track, would you be interested?” The fund I mentioned had just made a big investment in Stability AI, so we had some credibility there. This was very early in the current wave of AI, right as ChatGPT was coming out.

I’d read the network state book and was interested in those concepts, but I think it took that experience of actually being there — having the lived experience of it — for me to really understand why this kind of thing is so interesting and potentially so important.

4:38 Isabelle Castro

I’d like you to expand on that point — why do you think network states are important, and how has that informed your journey into creating Edge?

4:59 Timour Kosters

I think network states represent a very broad set of ideas, and like all good movements, they start to experience some semantic drift. I worked with DAOs for some time, and after a few years no one really knew what anyone meant by the word “DAO” anymore. Something similar is happening with the network state movement. But within that, there are some really interesting early glimmers of importance.

For me, the long-run vision that’s most compelling is this idea of international groups of people that transcend national borders, who have a sense of collective agency and collective purpose, and who use the latest technological tools and cultural frameworks to push those ideas forward.

The reason I think these things matter is that in the same way blockchains offer an experimentation ground for new incentive models, economies, and financial systems, I think network societies — network states, network cities, pick your nomenclature — offer experimentation grounds for new societal constructs. How do we learn? How do we teach our kids? How do we govern the communities we live in? Is it actually possible to create communities that transcend national boundaries but are also genuine sources of belonging — things people actually feel part of, and maybe eventually feel more kinship with than the place they were born?

Vitalik had an interesting post recently — I think it was called “Let a Thousand Societies Bloom” — and he addressed a common criticism of these experiments: that it’s just a bunch of rich people playing around, billionaires trying to create their own little utopias. I actually agree many of those criticisms are valid and there are a lot of different people and perspectives in this space. But his point was that maybe that’s actually good, because otherwise those people would just take the reins of power somewhere like the United States instead. We’re seeing rising authoritarianism around the world, and his argument was that we should have far more experimentation in these fields. The network state movement is almost like a meme you can organise around — a set of ideas you can go to people with and say, “Hey, this is what we’re trying to do and why.”

8:44 Isabelle Castro

I feel like Edge operates in a unique position within this. One thing that stood out to me in my couple of days there was this focus on collaboration across very different sectors — artists working alongside technologists, with genuine philosophical interrogation happening too. I remember sitting in on a debate about truth in the age of AI in Austin that shaped a lot of my thinking. With this diversity of disciplines, how far do you think that shapes the culture and contributes to building something that could become a long-term society?

10:06 Timour Kosters

It’s a really interesting question. When I think about culture, I think about the practices, the norms, and the ways that people relate to each other in community. We see a lot of these things emerge quite naturally, and we actively optimise the experience for emergence. We want to create an environment where interesting things bubble up from the ground up. We don’t want to be authoritarian in how we organise our events — we want things to happen naturally and see what novel phenomena come out of that. That’s a very specific design decision.

We also optimise for multidisciplinarity, as you mentioned. We have people from all different fields, different walks of life, different ages. So there’s a lot of potential for different perspectives. The trick as an organiser is to create an environment that allows those things to actually be expressed. Because if it’s just a conference, it doesn’t really matter how many amazing perspectives are in the room — if there’s just one keynote after another, maybe people meet by the snack cart, but there’s no real collision of ideas.

That’s something we think about a lot. We have events that focus on emergence rather than big keynote programming. It’s much more about various forms of groups that form naturally and are almost fractal versions of the bigger community — reflecting all the diversity and plurality that’s in there.

Something I think about a lot is how you could take the same dozen people and put them in the middle of a forest versus a massive conference hall versus Edge City, and they’ll have very different interactions. So we think a lot about this “container” idea — what is the container people are in, and what sides of themselves and what group dynamics can be created within it?

If culture is the thing that happens when people start interacting, and the ways they interact over time form into practices and norms — then we create the spaces where what emerges is fun, engaging, and actually leads to new ideas, startups, and projects.

15:15 Isabelle Castro

Through your experimentation, have any core tenets emerged that go towards the culture you foster at Edge?

15:21 Timour Kosters

Yes, definitely. And just touching on the point you made before — I do think all these different communities will each play their own role, and that’s actually a really good thing. Different groups will have very different vibes that will crystallise into cultures that really appeal to some people and not others.

What’s interesting is that there’s something like the Silicon Valley model happening here, where a large part of its success is put down to the movement of key employees between companies — trade secrets and ways of practice are quite liquid, and everyone upgrades over time. I think something similar is happening with various network states. We share community members, and there’s a very liquid exchange of practices — “oh, this is how they do their morning check-ins” — and I think that’s a really healthy thing.

In terms of tenets that have arrived over time: we’ve tried to be very hands-off, but as often happens with new communities or organisations, you eventually realise there’s space for a values definition exercise. We’ve landed on four key principles for the villages, and they serve both as organisational guides and as a signalling mechanism — people read them, and they self-select into the community or not.

The first is multidisciplinarity. We have people from AI, crypto, biotech, science, music, art, and culture. The key insight is that we’re trying to build the precursor to what Brian Eno would call a “scenius” — the collective genius of a scene. That happens from the proximity of high-agency people who are enjoying life together. They might have very different opinions but find real joy in spending time with each other, and that often leads to constructive energy.

The second principle is co-creation. We never want to be the ones dictating how the experience should be. Over 90% of the sessions at Edge are not organised by us or someone we nominate. It’s very much a co-created experience. This sounds like we’re opting out, but it actually takes a lot of work — you have to set up the structures so things can be self-serve. How do you set up venues so anyone can put a session on a shared calendar and arrive an hour later to find everything ready? That takes time and a culture of care. It also leads to things we could never have imagined — someone came and built an entire wooden structure, or a solar data centre. I would never have thought to build a solar data centre, but someone had the energy and agency to do it.

The third principle is defaulting to health. We try to think of health as a core part of the infrastructure, not just a nice-to-have. There’s daily movement, access to good food, things like Muay Thai classes in Thailand. The key insight is that feeling good is a core part of flourishing, and when people flourish, their ideas do too.

The fourth principle is multigenerationality. People bring their kids, their families, their parents, their grandparents — it has a genuine village feel. This is critical for two reasons. One, a lot of the world’s most interesting thinkers have kids and families and can’t just disappear for a month — we get amazing people who are finally able to go somewhere they find interesting and bring their families along. Two, it just changes the vibe. It’s hard for it to be a tech-bro echo chamber when there are kids running around. Yes, they sometimes make noise in a lecture and you have to figure that out, but all in all it’s a massive value-add.

22:52 Isabelle Castro

The multigenerational element is actually more representative of society as a whole, isn’t it?

23:01 Timour Kosters

One hundred percent. And I think that’s a key differentiator. Vitalik has a line that basically says a two-day conference is a break from your life, but a two-month thing becomes your life — it’s long enough that you actually have to figure out how to live in that context. A one-month event becomes a much closer reflection of what society actually is, versus a one- or two-day event.

23:41 Isabelle Castro

That resonates with me. The first in-person network state event I attended was Network Society Camp just outside Austin. I was expecting something like a crypto conference — tech-bro fest — but it wasn’t. There were a lot of families, and I remember being taken aback. There were conversations about what schooling looks like in a network state, and about how to build for the elderly — things that simply wouldn’t have come up without the presence of families. The multigenerational aspect completely shifts what gets discussed.

24:41 Timour Kosters

Exactly. And it points at something really important: representation matters. If you want to build a diverse, functioning society, you need different perspectives in the room related to all the different areas of society you care about. Children are a very important group — they’re about to inherit everything we’re clumsily trying to create.

What’s your take, from having been to and spoken with people across a bunch of these projects? What do you see in the different cultures and approaches?

25:39 Isabelle Castro

I feel like certain projects focus on culture more than others, and they’d benefit from collaboration with the ones that don’t. This was a big theme at Network School and at the Network State Conference — the idea of layer ones and layer twos collaborating to create different things. When I was at Network School, which I’d consider a layer one, they really benefited from people creating layer twos within the community to build culture. Without that, we were essentially just occupying a luxury hotel.

But that brings me to my next question. I know you’ve been doing pop-up experiences but are thinking about longer-term residencies — the Esmeralda collaboration near Healdsburg, and a potential residency in Bhutan. How is the longer-term element changing how you approach these things?

27:29 Timour Kosters

It’s definitely affecting our thinking. Something that’s happened in the last six to twelve months in this movement is a lot more focus on permanent hubs as a goal or end state. When I think about what we’re doing with our pop-up villages, I do see them as a step toward a network of hubs around the world — some temporary, happening once or twice a year, some permanent over the next couple of decades. These pop-ups are almost incubators for the communities that will eventually go on to live in those permanent places.

That vision is already changing where we choose to do pop-ups. It’s always needed to be an interesting place people want to go to, with strong local community partnerships and a clear reason for being there. But increasingly, the important question is: what’s the long-term benefit to us and to the place? Can we form meaningful partnerships with organisations and institutions there?

For Edge City India — which we’re doing in Q4 of this year — there are probably 30 viable locations, and one of the deciding factors will be where we can establish an interesting partnership with a long-term local real estate project. We always have many local partners and are in touch with local governments and innovation agencies. But it’s more about finding a physical place we can actually benefit through our work, and over time create a partnership like we’ve done with Esmeralda and are beginning to develop in Bhutan.

In the long run, it’s about creating a network where learnings from different sides of the world can inform how things are done elsewhere — this amazing global community.

30:41 Isabelle Castro

The integration of local communities brings me to another question. How do you go about integrating local people in a way that’s meaningful — where they’re actually part of the fabric rather than just providing services?

31:26 Timour Kosters

It takes a lot of work. It starts with finding really strong local partnerships — with communities, organisations, or institutions that already have a lot of trust, and who see the value in what we’re doing and are willing to really understand it. Because it’s still confusing and illegible to a lot of people: “Wait, you’re coming for a month? What are your intentions — is it a corporate retreat? A conference? How many people — 500? A thousand? That’s insane.”

Let me give you a concrete example. We recently did a pop-up in Patagonia, Argentina, in a town called San Martín de los Andes — a beautiful mountain town about an hour’s flight from Buenos Aires. We got connected there through a friend, Cello, who runs Lemon, one of the biggest crypto success stories out of Argentina. He’s from there and knows everyone. We said, “Hey Cello, we’re looking at doing something in Argentina — where do you think?” And he said, “You have to do San Martín. I’ll introduce you to everyone.”

He introduced us to a local cooperative called Bitkesma, and they instantly got what we were trying to do. They followed up with us — which is actually quite rare; usually we drive the conversation — asking, “When’s it happening? When are you coming?” They were excited mainly because we were bringing 500 incredible builders, thinkers, and researchers to their town for a month, especially during the off-season. They introduced us to every restaurant owner. We got them all onboarded onto crypto, got free venues, and the state’s innovation agency got incredibly involved: “Wait — people from Silicon Valley and around the world are coming to our town for a month? What do you need? Visas? Anything?” That was very much the positive tail end — it’s not always like that. Sometimes people just don’t care. In California, actually, it can be a bit NIMBYish.

But generally, it just takes work. You have to go there, spend time, meet people, have dinner with them. We did an event in Turkey three years ago and we’re still in constant touch with our partners there — purely as friends. They’re always asking, “When are you coming back?” Some of the most meaningful connections that come out of all of this.

36:00 Isabelle Castro

That’s really interesting. I got the sense in Healdsburg that locals were happy for the event to be there but weren’t super engaged. In San Martín, when you were onboarding restaurant owners to crypto — were they actually curious about the conversations and activations happening in the community?

36:56 Timour Kosters

We’ve tried a few different approaches. The challenge is that you want locals to participate if they want to, but we also have a fairly strict application process for regular attendees. The balance we’ve struck is having one day a week that’s open to locals — people can come and experience the programming, host sessions, things like that.

In Patagonia, that was Wednesday, and we held a local demo day. What was amazing was that people came and gave ten-minute demos on things like new farming equipment they’d built — completely different from the AI startups otherwise presenting. But two of the locals actually got investment funding from our community. People went to the local demo days, they engaged, we had translators there, some sessions were in English, a lot in Spanish — and there were real outcomes. It wasn’t just theatre. People genuinely loved that part of the event. We’re always trying to improve on it, but it’s not necessarily hard — it just requires thoughtfulness and work.

38:47 Isabelle Castro

On the topic of your strict application process and fee structure — that’s been part of the external criticism of the movement generally. You acknowledged at the beginning that these events can appeal to a certain economic bracket. Do you see a world where these communities become more accessible further down the line?

40:21 Timour Kosters

I definitely don’t think it’s a good thing that these are expensive events that are therefore exclusive. We care a lot about this — over a quarter of the people who come to Edge are on some kind of discount or scholarship, and some come fully comped.

That said, we structure our events through a nonprofit. We’re not trying to make institutional money on the events themselves. They’re structured to break even with a small margin for error. We have ticket sales, sponsorships, and partnerships to cover costs. For people who can’t afford it, we have a robust scholarship system, discounts, and a volunteering pathway — you get a free ticket if you volunteer forty hours in a week. That sounds like a lot, but a lot of the time you’re doing things you’d want to be doing anyway; your attention is just also on making sure things are flowing.

The reality is it’s a balancing act. California, where we do Edge Esmeralda, is just an expensive place. We try to price tickets very reasonably, but accommodation costs a lot regardless. And the funny thing is, I hear very different things from different people. Some say, “Edge is so expensive — crypto conferences are free.” Others, used to TED or Summit where it’s £10,000 for a day or two, say, “Wait, it’s only £2k for a month — what’s the catch?” We’re just trying to stay sustainable while we figure out longer-term models through the network’s startups, real estate plays, and so on.

44:05 Isabelle Castro

I think you guys really do approach it thoughtfully compared to other projects. But it is a valid criticism and I’ve seen some things that feel more elitist, which is fine if it appeals to a certain type of person — it just wouldn’t be my primary community. I have friends who are artists, activists, who look at this and think, “Why would I spend that money for a month?” I understand all those perspectives. I guess when anything’s new there’s a lot to work through, and economies of scale might address some of it over time.

We’re coming close to the end of our time. What role do you see Edge playing in the wider network state and technology ecosystem?

45:40 Timour Kosters

I think what we’ve demonstrated so far for the network state movement is that you really do need to start with real, in-person gatherings that bring people together — so they actually have a chance to meet, connect, and form the bonds that will make the bigger vision real. When the book first came out, it painted a picture where people would meet online, buy land, and move in together. What it missed was that you actually have to bring people together first, build that culture, and build a vibe that people can opt into.

I think we’ve done that fairly successfully. We’ll keep demonstrating what really good versions of these things can look like and keep getting better at them.

We’ve also already had some significant startup successes and projects come out of Edge — a few where we can genuinely say they would not exist without having come through us. That feels very exciting, and I think it’s something we can push the broader industry toward: these are not just fun places to spend time, but interesting places for new projects to be incubated.

In terms of the wider role for technology and society — my sense is that things will come out of these spaces that could not have come out of a Y Combinator or a B2B SaaS accelerator. It’s a very different context. You have a lot of high-agency people living together for an extended period, so there’s a different milieu from which ideas can spawn. Maybe they won’t be printing unicorns the way YC does, but I think they’ll lead to genuinely novel ideas in the minds of people who spend time in those places.

48:19 Isabelle Castro

I agree. Some of the things I’ve seen in workshops at these gatherings have been pretty novel. And just the fact that people are around each other for extended periods seems to be beneficial — they’ve got the space, they’ve got the time to collaborate and create. Thank you so much for taking the time to speak with me about all of this.

48:59 Timour Kosters

Thanks so much for having me.

49:01 Isabelle Castro

And I hope I get to pop in to Edge Esmeralda in Healdsburg — I’m not sure if I’ll be able to make it, but I really hope so. Thank you, and have a great rest of your week.

49:19 Timour Kosters

Thank you so much. Take care.

In the months since, I have thought a lot about that quote. I’ve seen it reflected in a lot of the discontent with current systems inside and outside the world of network states. People approach it differently - some attempt to provide alternatives that sit outside of mainstream governance on blockchains, others choose to focus on their immediate vicinity, building small scale communities of friends, building new real estate projects focused on walkability and nature.

So what would “better” look like to you if you woke up in a position to materially change things? Would you go into it full swing and attempt to implement systemic change? Or would you go about it incrementally?

This was the question that faced Octavio Sanchez, the chief political architect of Honduras’ controversial ZEDE law – a law that has allowed for the development of Prospera and its layer 2 network state Infinita City, on the island of Roatan, Honduras.

The idea and the idealist

You might recognise this train of thought from “Petri Dishes” , a piece I published a few weeks ago that looked at medieval charter cities and their potential for IRL governance experimentation. Now, I’m going to take you through the real world example of Sanchez and his attempt to implement this theory into practice.

Sanchez was young and idealistic when he came into office. Since his teenage years he had been percolating futuristic policy solutions, going so far as to write a science fiction book at age 16, imagining government bulletins from Honduras in 2050.

His core conviction was that Honduras’ poverty was a problem of institutions. “Many all over the world,” he told NPR in an interview from January 2013, “don’t understand we are poor not because we are dumb; we are poor because of institutional arrangements, not because of lack of capacity to imagine things.”

By 2002, while working under future President Porfirio Lobo Sosa when Lobo led the Honduran Congress, Sanchez had begun advancing the idea alongside American development consultant Mark Klugmann. Together they envisioned a Hong Kong-style zone within Honduras, a place where better rules could prove themselves before spreading outward. The idea simmered for nearly a decade, waiting for the right political moment.

That moment came unexpectedly. A 2009 constitutional crisis, in which the military exiled President Zelaya for attempting to extend his term, reshuffled the political deck entirely.

Lobo Sosa won the election that followed, and Sanchez was sworn in as his Chief of Staff in 2010, finally holding the executive leverage to turn decades of thinking into policy. He and his colleagues were facing the tall task of changing the trajectory of what was then described as a “failed state.”

A modern charter

In 2010, Sanchez and Lobo Sosa along with National Congress President (and future President) Juan Orlando Hernandez sat down in Miami with Paul Romer, a celebrated economist who had championed the idea of bringing back charter cities during a TED talk in 2009.

“The real challenge then, is to try to figure out how we can change rules [...] if we can find ways to give more choices to both, that will give us a set of rules for changing rules that get us out of traps.

If you try to change the rules in the nation, you can’t give some people a chance to hold back, see how things turn out, and let others zoom ahead and try the new rules. But cities give you this opportunity to create new places, with new rules that people can opt in to. And they’re large enough to get all of the benefits that we can have when millions of us work together under good rules.” Romer had said.

Shortly after, the administration presented the idea of Regiones Especiales de Desarrollo (Special Development Regions) or REDs, quickly followed by the introduction of a law in July 2011. REDs signed off significant governance authority in special zones to attract investment and growth, much like other charter cities which came before. They were to be overseen by a Transparency Committee, which Romer announced he would lead.

The dream of the REDs was, however short lived. Following the revelation that an agreement had been signed without the Committee’s knowledge, Romer and four others resigned. While others involved said the Transparency Committee was not yet officially appointed and therefore the move was justified, in October 2012, the Honduran Supreme Court deemed REDs unconstitutional, and they were no more.

As we now know, that was not the end of Honduras’ special economic zone development, nor of Sanchez’s determination. Dismissing the Supreme Court defeat as a matter of internal politics rather than genuine public opposition, he, along with jurists Carlos Pineda and Ebal Diaz pushed for a new constitutional amendment. The ZEDE (Zone for Employment and Economic Development), was adopted the following year along with a new oversight committee called CAMP - Comité para la Adopción de Mejores Prácticas (Committee for the Adoption of Best Practices) of which Carlos Pineda became the secretary. The committee consisted of 21 global members, 4 of which were Honduran.

Subscribe now

With renewed fervor, Future President Orlando Hernandez campaigned heavily on the ZEDEs’ benefits and in 2017 Prospera was born.

Despite the authorisation, ZEDEs have been mired by controversy throughout their existence, with Prospera in the middle. Repealing the ZEDE law became a focal point of President Xiomara Castro’s 2021 campaign and during her tenure were subject to a number of votes to repeal.

The 2025 Honduran election became a nail-biting peak, off which the future of ZEDEs and Prospera could topple. Ultimately, Trump-backed Nasry Asfura won the vote, officially embracing the ZEDE model in early 2026 and securing Prospera’s future.

Bringing in charter cities to the present day

Sanchez’s journey with the ZEDE law that would allow the development of charter cities is a useful case study in what it actually takes to implement the structure for cities experimental governance to prove their case. Although founders now say there is much more appetite from global leaders to embark on this process, Honduras’ swings between acceptance and rejection demonstrate the rocky ground they have to navigate.

“Part of what’s unique about the Honduras program is that it wasn’t outsiders. It was their initiative. They created the laws and Octavio was critical in all of that,” says Patri Friedman in a call with me on the way to the airport. Friedman is leading the Liberty Acceleration Summit from 26th-28th March (next week) in collaboration with Infinita City of which Sanchez will be a key speaker. After the summit, he is hosting a “founders retreat” for network state project leaders, a more closed event that founders can apply to join.

“I’m really excited for him to come to an event finally and get some credit,” Friedman continues. He sees the summit as a chance to “celebrate the beneficial elections, celebrate Octavio and bring together Latin American officials,” who are interested in Honduras’ journey with the charter city model and special economic zones.

The summit itself is particularly focused on Latin American development as a whole – “Socialism is receding in Latin America. But will this be another cycle of the same old swings between right and left, or is this the time for real, lasting change? Could Latin America even be leading the world in liberal governance? The ingredients for a New Latin American Liberty are there, and they’re coming together at this summit.” reads the event page.

Talks include discussions on the geopolitical climate in the region and mainly feature leaders from the charter cities and network states movements, although Peruvian economist, Hernando de Soto is earmarked for a conversation in the morning of the first day.

Whether Honduras ultimately proves to be a replicable model for governance reform or simply a story of one unusually persistent insider navigating a uniquely chaotic political environment remains an open question. The ZEDE experiment is still young, its long term impact on ordinary Hondurans still unclear.

What Sanchez’s journey does demonstrate is something more modest but arguably more useful, that structural reform of this kind doesn’t move on ideas alone. It moves on relationships, timing and an almost irrational willingness to start over.

The summit next week will frame Honduras’ journey with ZEDEs as the beginning of a Latin American moment. That may be optimistic. But then again, so was a sixteen year old in Honduras writing policy bulletins from the future.

This week’s newsletter was sponsored by Infinita City.

Food for thought

• If you’re free next week and fancy a trip down to Roatan, theres still a few spots left to attend the Liberty Acceleration Summit. Patri’s founders retreat which takes place after is also open to application – I’ve met some of the people going and it sounds like it’s going to be great. If you (like I) are not able to attend, Infinita City will be posting recordings of the talks from the summit shortly after.

• If you’re not aware of Paul Romer’s approach to charter cities, take a few minutes to listen to his TED talk from 2009, it explains the theoretical basis behind many of the network state approaches that are being taken today. If you have time to go into 2009 governance ideas a little more, Paul Collier’s TED talk on post-conflict recovery within nations goes even deeper.

Share

FAQ Section

Q: What is Honduras’ ZEDE law and how did it come about?

A: The ZEDE (Zone for Employment and Economic Development) law is a Honduran constitutional amendment that allows designated zones to operate under distinct governance structures, with the aim of attracting investment and testing alternative institutional arrangements. It was developed by Octavio Sanchez, who served as Chief of Staff under President Porfirio Lobo Sosa from 2010, after an earlier version — the Special Development Regions (REDs) law — was struck down by the Honduran Supreme Court as unconstitutional in 2012. Sanchez had been developing the underlying concept since the early 2000s alongside American development consultant Mark Klugmann, drawing on economist Paul Romer’s charter city framework. The ZEDE amendment was adopted in 2013 with a new oversight body called CAMP, consisting of 21 global members.

Q: What is Prospera and where is it located?

A: Prospera is a ZEDE located on the island of Roatan, Honduras. It launched in 2017 following the passage of the ZEDE constitutional amendment and operates with significant autonomy from standard Honduran law, including its own regulatory and legal frameworks. Infinita City is described as a “layer 2 network state” built within Prospera’s structure. Prospera has been consistently controversial: repealing the ZEDE law was a central plank of President Xiomara Castro’s 2021 campaign, and the zone’s legal future remained uncertain until the 2025 Honduran election, when Trump-backed Nasry Asfura won and officially embraced the ZEDE model in early 2026.

Q: Who is Octavio Sanchez and what role did he play in Honduras’ charter city experiment?

A: Octavio Sanchez is a Honduran politician and policy architect who served as Chief of Staff to President Porfirio Lobo Sosa and is credited as the primary driver of Honduras’ special economic zone legislation. His interest in institutional reform began in his teenage years — he wrote a political science fiction book at 16 imagining government bulletins from Honduras in 2050. His core argument, articulated in a 2013 NPR interview, was that “we are poor not because we are dumb; we are poor because of institutional arrangements.” He navigated two rounds of legislation, a Supreme Court defeat, multiple repeal attempts, and more than two decades of political turbulence to bring the ZEDE framework into existence.

Q: How do charter cities relate to the network state movement?

A: Charter cities and network states share a foundational premise that governance can and should be tested in bounded, opt-in environments before being adopted more broadly. Economist Paul Romer, whose 2009 TED talk influenced Honduras’ RED legislation, argued that cities allow new rules to be trialled without forcing an entire nation to adopt them simultaneously. Network states extend this logic into digital and community-first contexts, sometimes anchored on blockchains, sometimes in physical communities. Prospera’s relationship with Infinita City, described as a layer 2 network state built within the ZEDE, represents one attempt to layer these models. Patri Friedman, who is leading the Liberty Acceleration Summit in late March 2026 in collaboration with Infinita City, sees Honduras as evidence that governance experimentation can originate from within a country’s own political figures rather than being imposed externally.

“Would Christoph be up for a debate about the advantages of AIs listening continuously versus the possible breach of privacy and surveillence concerns?” I asked the PR who had contacted me.

“No problem at all!” came the response.

What followed was an in depth discussion on enterprise collaboration, privacy, secure context-aware AI and the limitations of current remote communication tools. The conversation also goes into virtual reality as a richer medium for building group culture in remote organisations.

If you’re thinking seriously about the future of work, enterprise AI, collaboration design and human-centered technology, this one is for you.

Share

Topics covered:

• The difference between passive surveillance and explicit AI participation

• Why context is the missing ingredient in useful AI

• How enterprises can use AI to surface hidden insight across teams

• Why VR may be more than a novelty for remote collaboration

• How presence, immersion and spatial interaction can change the dynamics of work
  

Transcript

0:00

Christoph Fleischmann

I mean, I think it would be really scary if you had this little icon in the top right of your laptop and it’s kind of listening and transcribing all of your life. But in in our approach, we we try to be very upfront and say, “Hey, this is your this is not just, you know, an AI that stands here in the room with you. That’s that’s your meeting assistant. That’s your facilitator.

~Intro~

0:34

Isabelle Castro

Hello and welcome to the Utopia in Beta podcast. I’m Isabelle Castro, your host, and today I’m joined by Christoph Fleischmann, founder and CEO of Arthur Technologies.

Hi Christoph. Great to have you on. How are you today?

1:11

Christoph Fleischmann

Hi Isabelle. Great to be here. Thank you so much. Doing excellent. How are you doing?

1:19

Isabelle Castro

Yeah, I’m doing well. Sunny day today. I’ve been looking forward to this conversation because when I was pitched Arthur as a topic, I was told that it has an AI that kind of listens beyond the standard kind of video calls and recordings like we’re creating now. And I’m keen to get into that whole thing as it brings up a whole load of questions for me. But before we do that, maybe you can tell me a bit about you and Arthur and how you came to be.

1:51

Christoph Fleischmann

Yeah, happy to. So I’m Christoph, founder and CEO here at Arthur. We focus on enterprise collaboration. We try to apply new technology to the challenges that big companies have when it comes to collaborating, ideating, making decisions or just breaking down silos. One of our products is actually an immersive product where people can wear VR headsets and meet no matter where they are in the world, as if they’re in the same room. And over the past couple of years, we’ve added more and more AI products to our portfolio — both in the VR world but also in other tools where we allow AI agents to facilitate communication in enterprises. We try to find a good balance between privacy, security, and utility. It’s very important to understand that you will want to have an incredibly relevant AI, but you need to make sure that it is secure, it is gapped and your data is protected. And even if the data itself is secure, you want to make sure that only the people in your team or organisation who should know this information actually get to see it.

3:12

Isabelle Castro

Okay, cool. So you do kind of go into the whole privacy and kind of information gap. That is a relief.

3:21

Christoph Fleischmann

Yeah, I think it’s really important. You know, us being an enterprise platform, I think consumer products very often have this challenge that you need to think about ads, about monetisation, and their user data is very important. For us, user data is what allows us to hopefully be a productive tool, but at the same time the biggest priority is to make sure that this data is never in the wrong hands.

3:51

Isabelle Castro

Okay, cool. The VR element is super interesting — I just wrote an article about it, and I will be going into that because it’s super exciting. But going back to this whole idea of AI that listens — how does this AI work? Can you explain a bit beyond the one-line pitch I got, which was “an AI that listens”?

4:21

Christoph Fleischmann

Happy to. Our product Arthur Vibe really allows you to have an assistant that can go out into your organisation and talk to people on your behalf. In many organisations, you might be facing the problem that there are too many people, too many stakeholders, and the topics are so complicated and complex that even if everyone were constantly in meetings, you could probably only cover a tiny fraction of all the information that’s actually there. And I’m sure everyone listening doesn’t want to hear the pitch that we need more meetings.

How do we solve this? One way is just to live with the fact that I might have an entirely different view on a problem than you, simply because we never had a chance to exchange our thoughts or concerns. Now imagine we’re in the same team and we have a manager — she might have 40 people whose perspectives she’d like to know on a particular problem.

With AI it becomes interesting, because until now we’ve been able to use digital tools to visualise information — dashboards, charts — and this has been around for 20 or 30 years. But it still doesn’t really relay nuances or give you perspectives. And what happens is that a lot of the time the people who might be the most interesting to listen to are the ones who never speak up — their ideas never break through the surface.

So we took a very tech-positive approach and asked: can AI actually help here? Can I use a tool that has the time to have an intelligent conversation with all 100 people involved in a complicated process? Can I design it so that everyone can share their ideas and thoughts in depth with a digital assistant? And can I then build intelligent pipelines that allow us to analyse this and extract information from it? That’s really what we’ve built with Vibe.

In some enterprises, you would hire a whole team to go out and interview people for weeks — and this just makes the process so inaccessible because it’s so expensive. We thought: what if we could give you a tool where you basically tell the AI, “Hey, I have this problem. Things are not going well in this process and I can’t put my finger on it. I hear all these different perspectives. I just need you to go out there and run a proper analysis. Talk to everyone and let me know what’s going on.”

And that’s really what we’ve built. You configure it, it goes out, individual people get a short email, they open it, and it looks like a video conference — an AI greets them, introduces the topic, and has a natural conversation about it. Maybe five minutes, maybe ten. If you want to go further, you can talk for 20 minutes. All of this information gets saved into a central knowledge database, because we want the little nuances not to get lost — that’s what makes it really helpful and productive.

As a manager, what you then get is both a report on the problem or topic — not always a problem, maybe just a place to hear good ideas — and an aggregated view that is balanced and readable. But if there’s something you’re really interested in, the report is interactive. You can double-click on it and ask the AI, “What did Isabelle mean when she mentioned that?” And the AI thinks for a second and says, “She was talking from this perspective, which also John and Maria agree is a problem.”

This is how we’re trying to rethink collaboration in the enterprise — where the idea of synchronous collaboration, where we’re all in meetings at the same time, is beautiful and I love being in the same room with people. But it’s just not always possible, and we still want this important information not to get lost.

9:26

Isabelle Castro

Okay. So it’s not like it’s listening constantly or tracking constantly what each employee is doing. They click, they have an interview, and then they can go off and it won’t be tracking anymore. It’s very explicit — you really know you’re talking to an AI and sharing information.

9:43

Christoph Fleischmann

Exactly. You get very specific questions about the problem. Now that’s our app Vibe. We do also have, in our VR meeting app, an AI that can be there in a workshop and kind of listen in — but it’s like a participant. You’re very aware that it’s there next to you. So I’d again say it’s an interesting user experience question.

I mean, I think it would be really scary if you had this little icon in the top right of your laptop that’s kind of listening and transcribing all of your life. But in our approach, we try to be very upfront and say, “Hey, this is not just an AI that stands in the room with you. That’s your meeting assistant. That’s your facilitator, maybe even.” And this way I think it becomes a little bit more natural.

What we do in both products is spend a lot of time thinking about the security architecture, the access to this information, and what we do with it. We don’t save specific data about the person who said these words — we save the ideas and try to connect them to a knowledge graph so our AI can learn.

And here I do want to bring up why I actually think it’s good to have AI that’s not just receiving documents or little queries via chat. You can make the point that you actually do want it, in specific settings where it’s clear, to be there listening to you. Because — and I’m sure you’ve seen it too — it’s amazing how quickly we can produce content with AI, but the idea of “AI slop” is very prevalent nowadays. It’s this low-effort, high-quantity stuff. How do you get over this hump where everyone can now write you a 50-bullet-point email with ideas, and all 50 are probably useless if the initial prompt was just “come up with 50 ideas”? It’s the missing context that makes AI sloppy.

And how do we get context? If the AI is in the meeting, if it hears you emphatically arguing for an idea, it can save this information and learn about it. And then if you task it later to help you with something, it can remember. That’s such a different view from just receiving a spreadsheet and a question. Such a different view from saying, “Knowing everything you’ve learned — from our asynchronous tool Vibe where 100 people talked to you about their ideas, or from our synchronous VR meeting where 10 people had an engaging workshop — now help me.” We just think we can build actually useful AI this way. We need to meet people a little bit more where they are.

13:17

Isabelle Castro

Yeah, I mean, AI transcription tools have kind of changed my life. I remember when I first started in journalism, I would transcribe myself and it would take up most of my time — AI has made that a lot easier. What I have noticed though is that sometimes it misses the point and the nuance of what’s being said — it highlights things it thinks are important but actually aren’t the key theme of the conversation. How do you address that? How does the decision-making work in the AI for it to bring up and highlight the main points?

14:12

Christoph Fleischmann

Accuracy is just a really hard problem. I don’t think there’s a single silver bullet. A lot of things need to go right for an AI to actually not just sound smart, but actually be smart if you’re an expert in the area.

Our approach is to spend a lot of time trying to give this AI as much context as possible — context about your work, about your priorities. Always being very careful to focus it on the subject matter rather than people and personal information. What are the products in your company? Who is involved? What are the challenges? The more information we can give the AI about these priorities, the more intelligent it’s going to become. But it’s still not a guarantee it’ll always get the right point. I think this is where we still notice it is not human — it is still mathematical calculations happening at such a large scale that something seemingly intelligent comes out.

You’re certainly right that this is one of the biggest challenges. Once you add the right context, you start reaching a point where you really have to search for where it gets things wrong. And in those cases, the only solution I believe is to make it as explicit, transparent, and easy to navigate for the users as possible.

The worst thing I can imagine is speaking all day into a black box and never hearing anything back. It would be so much better if you could chat with it. So one of the features we’ve built is that if the AI thinks it has conflicting information — if someone says something that might be logically contradictory — it might ask a clarifying question and say, “Can you let me know which one of these is true, or how did I misunderstand that?” Every human would naturally do this. AIs don’t. So you have to programme them to find contradictions and update their knowledge. It’s a lot about context management and how we interface with these systems. If we make those two things easy, accessible, transparent and secure, I think we can build really helpful AI that truly drives us forward instead of just creating a mess for us all.

17:17

Isabelle Castro

And so the data it’s kind of trained on — is it things that are inputted specifically for each enterprise, and then on top of that it adds information from interviews and starts learning beyond that? And it doesn’t outsource research capabilities to the general internet? Or is there some kind of integration with the general internet and Reddit and all that kind of stuff?

17:58

Christoph Fleischmann

We do have features where we actually want the AI to break out of the knowledge silo that we have — but again, it should be explicit: “I’m doing a web search. I’m going out there. I’m trying to get the latest headlines for you.” But you’re absolutely right that the way we’ve built it, you have a foundation of knowledge and you keep adding to it. It’s very interesting how it develops — it’s a knowledge graph. It starts identifying different topics that it sees recurring and tries to build a model of understanding of how your team works. We’ve built that on an individual level, a team level, and an organisation level — and it understands the differences.

We give the system different levels of access. You as a user primarily have access to your own data or information. And if you choose to share or make something available to others, you have pretty explicit controls about that.

19:13

Isabelle Castro

Okay, that’s cool. And I mean, I can imagine this being integrated for a year in a business — that’s a lot of information about the business and probably really valuable to hack. On the security front, how do you protect against those kinds of risks?

19:39

Christoph Fleischmann

I think there is already a lot of very sensitive data that we store as enterprises in a lot of software — even as consumer users, we share some of our most personal things in our chats. So we live in a world where we’ve made the decision to trust some providers to be good stewards of our data. In the consumer world, this doesn’t always go right. In the enterprise world, because security is such a centrepiece of why you’d buy software specifically for a company, it’s more built into the logic. You have no choice but to architect it to be secure in the first place.

Even if we deal with a lot of data, it’s pretty much the same as saving spreadsheets to Google Drive or Microsoft SharePoint. We’re still storing information on servers — that’s not different with AI. Technically the information is stored in a different format, but from a security aspect, storage may not be so different from what we’ve been doing for 20 or 30 years, where we have processes and standards in place.

It becomes more interesting when you ask: can you trick an AI into giving you data it shouldn’t? That’s a topic where you obviously need to work very closely with the providers of these large language models, who spend a good amount of their time on security — specifically ensuring the AIs themselves don’t say something wrong or give out the wrong information. And for us, in every feature we build, one of the first questions we have to ask is: how does this affect security and privacy? If you make that a habit, I think you can build very safe systems for your users.

22:18

Isabelle Castro

Interesting. I was talking to my friend the other day and his business uses ChatGPT Enterprise, and he realised that he has access to everyone else’s chats — and he’s not an overarching manager. He can see some of other people’s queries, and that was a realisation for me. I didn’t think that was possible. Because, like you said, I would imagine that for these enterprise software solutions, security would be front of mind. There’s a very fine line between helping a worker or a team and... spying.

23:05

Christoph Fleischmann

I mean, that does sound scary. I haven’t seen this in the enterprise lineup from OpenAI — I’m sure there is a setting where you can share chats, but if it’s accidentally set to “on” by default... yeah, you really have to make it absolutely clear whether a chat is private or not. I can’t think of cases where for us this would be an issue — we don’t really have group chats where you might be accidentally sharing information. So I don’t think this specific issue is something we’ve encountered. But wow — what a great example. Hopefully he hadn’t uploaded anything too embarrassing!

24:17

Isabelle Castro

Ha! No, I mean, he was saying things like “how do I tell my boss that I don’t like him?” But the ones he was seeing from others were mainly kind of like “how do I make this sound good?” and then something pretty rough being used for a proper presentation. But yeah — good to know you haven’t had any instances of that. Have you looked into decentralised storage? I write a lot about that and how it can be better on the security front — like having a distributed system so you don’t have one attack surface. Is that something you’re looking into or have you explored it and found it’s not a good option?

25:07

Christoph Fleischmann

Yeah, so in our case it’s interesting on two levels. We very often work with established cloud providers who I believe already employ a lot of sharding principles by design in their architecture. One of the things for startups, of course, is that you don’t want to be in a position where you have to solve every problem yourself. You need partners — the Microsoft Azures, the Amazon Web Services, the Google Cloud Platforms of the world. They do a pretty good job at handling data-level attack surfaces.

I find it fascinating when thinking about splitting or sharding data in the context of AI use cases — because obviously a big part would say it’s so easy if you dump everything into the same place and the AI can read through everything. But I imagine the future is going to be a lot more dynamic. An AI will make decisions about where to search for data and will probably talk to some other intelligence system that will very quickly decide whether it can have access to that data or not. These two different systems working together.

I do think we won’t store thousands of copies of these blocks everywhere, but we will probably have some very intelligent and secure way to distribute this data — and maybe by distributing it, make it very hard to decipher if you just have one puzzle piece. It’s something we’re not fully familiar with yet, to be honest, but I think it’s going to be very interesting as we see the build-out of more and more AI infrastructure.

27:18

Isabelle Castro

No, for sure. So I just want to kind of look at the use case for this. Does it work for all kinds of teams? Does it work for creative teams, for example — which is a slightly different approach to, say, a marketing or sales team?

27:39

Christoph Fleischmann

With a tool like ours, we’re actually in the difficult spot that you could try to deploy it in so many different ways. I think it’s good to pick a few at the beginning and get really good at them. Ironically, the kind of use cases we see people go for are the very opposite of the creative one you just described. A typical case is an incredibly dry, complicated 80-step process internally in a supply chain where there are legal standards to keep in mind, internal documentation standards, and it becomes this completely non-human-understandable process. Or in a company of 10,000 people, you have only one person who’s been there for 40 years who maybe understands the whole thing.

This is where we see it really shine — because it’s able to understand this, create abstraction layers of a very complicated process, and then talk to individual people about components of it. It acts like an investigator to help you understand, not to find an individual to blame, but to find where the system has a little gap or a problem.

I would love to see whether it can help with creative writing. I would love to see whether there’s value in synthesising ideas from 50 or 100 people — or whether that’s actually where you want a singular individual to be in the zone and come up with something. Something I’ll definitely go back to my team and try.

29:58

Isabelle Castro

Yeah, it’d be interesting. I’ve mainly worked in creative teams before — journalism, art and art projects. So this is an area where I’m used to quite a lot of meandering. If an AI were to come in and interject and say, “Oh, you’re having opposite views,” I’d be kind of like, “Okay, just chill. Let’s finish that thought first.”

But it would be interesting to see how it would interact with that. Could it create a kind of central oracle? Something I look into quite a lot with my writing is truth — how one creates some kind of truth, maybe even using AI to collect a lot of data and a lot of opinions. And it sounds like Arthur could potentially do that in a hypothetical situation where you’re not focusing on enterprises and the business use case. Where do you see something like Arthur being useful hypothetically?

31:20

Christoph Fleischmann

I certainly think that when it comes to innovation, research, and really being at the frontier where you want people to come up with great thoughts and exchange them easily — creative fields and the arts could be incredible.

I’m such a fan of human art that I’m having trouble imagining AI playing a bigger role than just a facilitator of communication, honestly. And to be honest, that’s the same role we want it to play even in our dry enterprise use cases. We don’t want it to come up with its own idea for your business — that’s not going to make sense. But: help me get information from 50 people, 49 of whom I’ve never actually heard speak about the problem. Help me get this information and then let me talk to you to understand what they’re saying.

I do think it could be interesting in creative contexts. Imagine you want to build a large, complex fantasy world for a novel — it could help you keep track of different characters and roles and storylines. Anything that has complexity to it, I think we could deploy it there. Not to replace anything we’d do as humans, but to augment us: “What did I say again that this character did?” Maybe that’s how I could see it.

33:17

Isabelle Castro

Yeah. As you talk, I’m kind of thinking — maybe because my head is tuned into new governance systems — but this could be helpful in the context of a member of parliament trying to converse with their constituents. That would be a really interesting application.

33:44

Christoph Fleischmann

I’d love it if we did that. We do work with some governments and entities that deal with the challenges that anything public faces — the divide between the government, which by default is the most complicated, largest organisation because it involves everyone of all different backgrounds and needs. We’ve seen in our VR world that we can help municipalities, for example, reach their constituents more easily. And with Vibe, this aggregator of information, there are certainly things going exactly in this direction — though I can’t speak too freely about it yet. We can say: can we get richer data? We don’t just want numbers. We want people to tell us more about what they need.

I’m just so excited to work on this because if we do it right, it’s going to help drive human collaboration so much further. I see so much inefficiency in the world — which is beautiful in some way, but it stops us from achieving a scientific breakthrough or finding a better way to bring our product to people. I’m really excited about what AI can do there. I’m super not excited about AI writing emails — I write every one of my emails because I like to be in control of how I convey my thoughts. But if AI can help me aggregate data and make a better decision because it involves more people’s thoughts? Wow, I think that would be great.

35:57

Isabelle Castro

Yeah, that is an exciting area. The future is exciting. You mentioned your VR rooms — that’s something really interesting to me. How has adoption been on that? Where do you see more people using it? I’ve been a remote worker for a lot of my career. Is it mainly remote workers? How does it work?

36:37

Christoph Fleischmann

I think it’s the ultimate medium for how we can consume digital information. There’s no way richer than to put on something that immerses you fully in the world. We’ve had so many sessions with clients where all you have to do is put on this headset — and even though it’s technically like a video game, a 3D animation in front of you, you just feel like you’re there with the other person. You hear people from where they’re coming, and something in our brains switches on when you are in a 3D room rather than looking at a flat video. I’m no neuroscientist, but I’m very grateful for the folks who are looking into this. For me it’s the richest medium, and in a way it’s inevitable that it will be an important part of how we connect.

I think we will want thinner headsets — ideally like sunglasses, really comfortable. Because they’re not that comfortable today, we just need to pick use cases where it really makes sense. For an interview like this, it’s probably overkill. Or if I just want to call my mum and chat about our days, I probably don’t want to put on a VR headset.

But if I’m running a training with eight people that I cannot bring into the same room — they might be in different countries — it becomes incredibly powerful because I can craft a very immersive, hands-on, interactive experience where people really feel present. It’s the one thing everyone tells us: you are 100% locked in in these meetings.

We’re a remote company and we’ve always been remote, but we do every single one of our team meetings every week in VR. And I genuinely think this has had a huge impact on our culture. I genuinely think it’s different from an all-hands video call. You just really feel like you’re there once a week in this space that you call your little town hall.

I’m very excited about future generations of hardware that are coming — it’s going to break out from something you use very specifically for a complicated or important use case, towards something much more ubiquitous. And don’t get me started once we drop AI into these 3D spaces — suddenly you can chat on eye level with an AI that can assume a role and run a training with you. There’s incredible things to do in this whole area, and in the next couple of years the things we already know are coming are going to be really exciting.

40:10

Isabelle Castro

Yeah. There’s something about the 3D element, right — I guess it’s hardwired into our brains. From your own use of the VR element in your teams, do you get a sense that it creates more of a company culture? Part of being in an office is the chats next to the water cooler, the little jokes. Do you see a future where it’s not just meetings — where you have this whole office and you’re just kind of there? And do you see in your own use that it adds to company culture?

40:53

Christoph Fleischmann

I think this idea of the metaverse — honestly I was never a fan of the term itself because it felt a little inauthentic as a marketing term. But I think it’s inevitable that this medium is going to play a huge role in how we connect with somebody who’s not in the same room with us. And I refuse to believe that humanity is going to say, “Okay, we have an HD video call and this is the peak of what we can pull off in terms of connecting over longer distances.” It’s not going to be like this. We’re going to have very light, comfortable headsets, and I think what we’ll be able to create is a way to connect that doesn’t have to be exactly like the real world, but is able to mimic so many of the things that are important to us as humans.

This feeling of presence is hard to describe. You ask me about our company culture — I really look forward every week to logging into this meeting. I walk around at the beginning and go to the first group of folks. I chat for a few moments: “How are you guys doing?” And then somebody else, I hear their voice in the back: “Hey Brian, how are you? How much snow do you have?” We’re definitely early adopters — weirdos, in a way. We have people in WeWorks putting on these VR headsets and everybody’s staring at them.

But I think it’s something very, very powerful. And I think it’s not dystopian — because what we allow people in our company is to prioritise where they want to live their private lives and not make a compromise on their career. In a world where I can give my employees a choice, I will always say: “Pick where you want to live physically, but then meet me in this world so we can still have at least something like being in the same room.”

43:38

Isabelle Castro

Yeah, that’s really really cool. I was imagining that you were all just sitting around a boardroom table, but the fact that people can come up behind you and you can actually be like, “Oh hey” — and like a holiday party where somebody brings a reindeer model and music and... that’s super weird but it’s real.

44:06

Christoph Fleischmann

That’s so cool. Yeah, we do our holiday parties in virtual reality. It’s super weird, but it’s real.

44:25

Isabelle Castro

That’s so cool. If you’ve got a video of that, I’d love to see it. Okay, well, I think we’ve come to the end of our time. Thank you very much for coming on — I’ve really enjoyed this conversation. I came to it kind of scared of an AI that listens all the time, and I’m going away from it very excited about the fact that maybe we’ll have virtual worlds for the workplace. Thank you, and have a great rest of your day.

44:53

Christoph Fleischmann

It was a pleasure. Thank you as well. You too.

And yes I realise how vague that is, I could be writing about any manner of things related to AI or even politics for that matter when talking about “the Machine” but this is a specific one.

In 1909, E.M Forster wrote a short story titled, The Machine Stops which tells of a future where all humans are living underground due to degradation of the natural world on the surface. They are pretty happy. They live alone in hexagonal rooms, and “the Machine” takes care of everything. If they want to sleep, they press a button and a bed pops out. If they want to eat, same thing. Their contact with other humans happens through videocalls and holograms and they dedicate themselves to “ideas” – sharing them, going to lectures and talking about them – if an idea isn’t present, the interaction isn’t worthwhile (therefore one character’s venture to the surface and experiencing grass for the first time was somewhat boring, as there was perceived to be no idea behind the experience). Their dependency is so embedded that when the Machine starts to fail, they can no longer function and eventually everyone dies - cheerful.

My encounter with the machine wasn’t nearly as dramatic. It came from my first in-depth session building with Claude Code. Of course, I’ve heard about what Claude Code can do – its “magic” – from those around me, increasingly so in the last few weeks, so I thought I’d give it a go.

Freelance writing comes with a lot of benefits. I can choose my schedule and location, turning up to any event or network state around the world (visa permitting) at a moment’s notice. But I do often long for the back and forth of ideas within a traditional newsroom, as well as the administrative support. So I thought I’d try and build it with AI agents.

Share

I’m not going to go into the process here, unlike others. Maybe I will do another day if people are interested BUT what I will say is I’m impressed. Within a morning, I had set up four separate agents, one to log my ideas and organise them into a publishing calendar, one to act as a sounding board for new ideas, one editor that will bluntly rip my first drafts to shreds and correct my grammar, and one that scrapes my latest Substack post and generates a few ideas for social media (I tried to get it to generate social media posts outright and, as ever, was disappointed at the result).

Finally I could see what all the fuss was about – why subscribe to a whole load of different services when you could just build them and power them with one app?... That, there, was my glimpse of the Machine.

Suddenly a future where one operating system powers the whole world didn’t seem that far away.

The Holy Grail of app development

One thing has remained consistent in the five years I’ve been writing about tech – the march towards the holy grail of online services - “the everything app”.

Revolut has been building that way for a while, evolving from a simple currency exchange app to now hosting an accommodation booking app, access to airport lounges, global “experiences”, as well as all the financial functions you would expect from a neobank in 2026.

Google and Amazon, I guess you’d call them more service providers than apps, have continued their steady take over of the internet. Meta too, despite being blocked on the attempted leap to financial services in 2019-2022 with Libra/Diem (a direction that seems increasingly possible in the foreseeable future). Telegram has its mini app ecosystem and a wallet. Even Elon Musk has been very clear about his intentions for X to follow a similar trajectory.

But most, so far, have failed to do “everything” – in fact the only app that really has succeeded is WeChat, which has 1.4 bn monthly active users and processes 40% of China’s mobile payments, despite starting life as a messaging app. It covers the payments, social and gaming needs of its billions of users, while offering “WeChat mini programs” allowing businesses to plug in and provide services via the WeChat interface. A restaurant using WeChat can use mini programs for ordering and delivery, payments and loyalty points, and often do.

For customers, the use of a single app is appealing, it makes everything just that little bit more convenient – that kind of convenience you don’t need until you have it, and then you just can’t do without. I remember during my brief stint living in China, it got to the point where I would become frustrated if I happened to dine in a place which didn’t have the WeChat pay option. People are genuinely exhausted from managing SO MANY SUBSCRIPTIONS, so much so that businesses are popping up to help users keep track of them in one place.

Plus, integrating everything makes certain services so much more useful – of course I want Google to have access to my Luma and Partiful. How else am I going to remember the events I wanted to go to?

For the app providers, it’s a flywheel. The increase in services edges the customers’ switching cost just that little bit higher, increasing stickiness and enhancing the potential entry points for new users.

It’s the most defensible business model ever built, with every added service deepening the moat.

Big boots

Going back to Meta.

Meta owns social media. With over 100 subsidiaries, including Instagram, Whatsapp, gaming platforms and the VR/AR tech to make video sharing just that bit easier, it’s got the social stuff of the present day, as well as the future, sorted.

But when, in 2019 they started a project to create a global stablecoin payments system (Libra, later renamed Diem), jumping into the mobile payments game, they were met with significant regulatory backlash.

Why? It was deemed anticompetitive in some Western regions, while in others it was called a threat to the existing banking system due to the global reach of the company, so similar reasoning. Aside from that, we were still in the times of debate around whether a stablecoin is actually money.

Meta is not alone, many of the big tech players are regularly subject to antitrust lawsuits – if they reach too far beyond their boundaries they are subject to some sort of scrutiny. This has left the playing field open, albeit marginally.

Now I’m not including all this as a judgement of big tech acquisitions, as one builder told me a couple of years ago, they provide a lot of value for start-up founders. In fact, many are building for that outcome.

But the era of the everything app and the single subscription economy may be rapidly approaching, powered, ever enthusiastically, by AI.

The single subscription

“You’ve just got to decide whether you want to keep renting software, or spend an hour, once, to create something you actually own.” It’s Sara again - she’s filming a video for a sponsor who recently launched a feature to vibecode phone apps. She had just made a real time translator and pronunciation tutor app in just over a day, so she had a point— it’s what really pushed me to create something of my own.

Two days later I’m logging onto Claude and discussing how to solve my lack of newsroom blues.

As someone who has never built anything beyond an interactive art installation, the process was pretty magical, we had a chat, it told me what to do, and if I came up against any problems (like computer storage) it found a way around it.

I saw the reality of what this could be, that people would just be able to whip up something to solve whatever that niggling problem they had. Later that day, in a chat with my uber driver, I found that it was already having an impact. Through using AI, he told me he had been able to create two side hustles, both of which worked while he was on his Uber run. “I’m on track to make six figures from both of them this year, just because of AI,” he said — it was a refreshing change from the doomerism I’d been reading online.

While writing this essay, I took a short break to interview a founder building financial tools for AI agents who told me - “People are now using agents to build businesses while sitting on the toilet, or on the bus to work… all it takes is 30 minutes now and they have them working to build it in the background.”

But through the process of building my agents I did come across a realisation – as someone who had never built anything, I didn’t know any better. Claude was telling me it needed to connect to a specific API and highlighted the one which would be easier for it to set up, so I went with that one.

I’m like the majority of people who will be using vibecoding apps. Fundamentally at the whim of where it wants to take me. Regardless of why it’s directing me in a certain direction, I’m happy to go that way. All I wanted to do was make my content production pipeline better anyway.

And yes, there are the open source platforms that allow for self hosting but the average internet user, newbies to the coding space without much incentive to spend time going into it deeply, are infinitely more likely to just go with the shiny all-in-one platforms that they heard about down the pub (or during the Superbowl).

You see it in other AI usage too. The use of search engines, for example, is rapidly declining — people are just happy to chat to their Claude, Gemini or (less so now due to recent events) ChatGPT, to find out an answer. You almost can’t avoid it - Google now has a whole load of AI summary and sponsored links to wade through before you can get to normal website discovery . This shift is already impacting how people market themselves online. Platforms like OpenAI are selling advertising, which will inevitably slip results and links ever so subtly into its sycophantic wordplay. SEO is changing to ASO (agent search optimisation) – the focus is no longer on how a user finds you, it’s more about discoverability to the AI.

Going beyond pay for play, agents have an incentive to use the services built to interact with the specific platform they are built on. It was suggested that I build my agents on Google Colab, so the APIs it plugged in, Claude explained, would be more seamless and easier to operate if they came from the same developer.

If agents become the primary method for interacting with the internet, we wouldn’t even know if a platform was really overstepping their antitrust boundaries. It’s all in the background. In this scenario, would we even care?

Back to Sara’s point - “do you want to keep renting or build something you own?” I’m on board with the ownership part, all for the “Own” era of the internet, but what we start to see in this reality is a shift to a single software subscription - your AI coding platform - regardless of whether you decide to self host the agent itself.

On top of that is the micropayment infrastructure that, again, your agent manages and has been designed by your AI platform following whatever incentives or interests that company happens to have.

“O Machine!”

The thing with the citizens of the Machine is that they were happy - they had everything at the click of a button. Everything was optimised and mechanical, to the point of losing many of the needs we take for granted now, like in person interaction. But they also didn’t have much choice.

And I’m not saying we are on the verge of living in single person pods, nowhere near it. But I do want to ruminate a little on what this shift to AI platforms being the main interaction point with the internet might mean.

At its centre is a double edged sword. On one hand, internet users are empowered to build whatever they want, solving all their digital needs. On the other, the tendency towards convenience, again, pushes power into a few well funded hands. And, along with it, users’ agency over choice.

The cost of switching, already pretty high within the ecosystem of established players like Google today, becomes infinitely higher. Not only does a consumer have the convenience of everything integrated into one place, but the AI is building a history of interactions, learning preferences, becoming the ultimate copilot.

In the recent exodus from ChatGPT to Claude, the cost of switching was tested. Anthropic CEO, Dario Amodei, took to X to tweet some helpful instructions on how to export the chat history with ChatGPT and import it into Claude. It was pretty straight forward, with OpenAI taking around a day to send over a download link to export the data. ChatGPT suffered a 295% surge in uninstalls, following the news around deals with the Pentagon.

The memory is heavy though, already packed with insights and preferences and we are only really at the beginning of the AI takeover. What happens a few years down the line?

In the blockchain space, builders and users are approaching data ownership, building infrastructure for people to own their data from day one. For many the answer is just buying an external server to self host, I’m just not sure that’s something the majority will do, despite hoping that it will be.

Even beyond the choice over what provider is the main platform, is the choice over what it plugs into. The agent provides a few curated choices and you make a decision. That could be a decision over the next API to plug in or the hotel you want to book with your booking agent – long gone are the days of 100 search results, we’ve whittled it down to just three.

In the world of Forster, the machine becomes the standard to live by - it’s optimised, logical, everything just works.. Until it doesn’t – which leaves the citizens unable to survive. Even in the run up to the Machine’s failure they make compromises in their own lives to allow for its errors, bought into the idea that it will just solve itself. The optimisation and logic of the machine becomes the north star - the thing that humans measure themselves against.

“There will come a generation that had got beyond facts, beyond impressions, a generation absolutely colourless, a generation ‘seraphically free From taint of personality,’ which will see the French Revolution not as it happened, nor as they would like it to have happened, but as it would have happened, had it taken place in the days of the Machine.”

In Forster’s story, the citizens had the capacity for something else and chose convenience instead, repeatedly, until the choice was gone.

It doesn’t need to get to that point. The “magic” is real – I built a solo writer newsroom in a morning. But so is the growing single platform reliance. Dressed up in convenience, it’s made to feel like a tomorrow problem. That’s until the platform does something that you don’t like, but by then, the choice may already be gone.

Food for thought

I’ve linked to a few articles and the EM Forster story in copy so I’ll only include a few links here today.

• Here is an interesting essay on how to restructure the economy to align with the agentic future. It covers the future of work, ideation and agency, but puts it firmly in the wheelhouse of the agents themselves, assuming that idea generation will be offset to the agents themselves.

• On the antitrust front, already people are seeing the limitations of BigTech antitrust litigation in the courtroom.

• Somewhat related to the idea of AI’s impact on choice, The Carnegie Endowment released a paper earlier this year about how AI might impact democracy, mapping the intersection of AI influence on individual choice as well as some opportunities. Related, I’m surfacing my chat with Lane Rettig, where we go into the idea of AI agent proxy voting.

• I wanted to go into the subscription state idea in the network state movement as a follow on from this, but already the essay was long. I did go into the idea briefly with Dan Thompson, in my latest podcast episode, but will likely do a dedicated essay shortly.
  

  FAQ SECTION

  Q: What is vibe coding and how does it work for non-developers?
  A: Vibe coding refers to building functional software through conversational prompts with AI tools like Claude Code, rather than writing code manually. A user describes what they want, the AI builds it, and troubleshoots problems in real time. Isabelle Castro, who had never built software beyond an art installation, used the approach to deploy four separate AI agents in a single morning — covering editorial organisation, idea generation, draft editing, and social media ideation — with no prior coding knowledge required.

  Q: Are AI coding tools actually replacing SaaS subscriptions?
  A: The argument is gaining traction among builders and creators. The logic is straightforward: if an AI can build a custom tool tailored to your exact workflow in an afternoon, the case for paying monthly fees across multiple SaaS platforms weakens. The shift is not yet mainstream, but the pattern mirrors what happened with search — people didn’t abandon Google intentionally, but AI chat interfaces gradually absorbed the behaviour. The same substitution dynamic is emerging with productivity software.

  Q: Why did Meta’s Libra/Diem payments project fail, and what does that have to do with everything apps?
  A: Meta launched Libra in 2019 as a global stablecoin payments system, later renamed Diem, intending to expand from social media into mobile payments. Regulators in multiple Western jurisdictions blocked it — citing anticompetitive concerns and the systemic risk posed by a company with Meta’s global reach entering the banking space. The project was abandoned by 2022. It’s a direct example of how regulatory pressure has, so far, kept the Western everything app model fragmented — a dynamic that AI-powered platforms may eventually route around, precisely because the activity happens in the background.

  Q: What is agent search optimisation (ASO) and how is it different from SEO?
  A: Traditional SEO focused on making content discoverable to human users through search engines. ASO — agent search optimisation — is the emerging practice of making content, products, or services discoverable to AI agents, which are increasingly the intermediaries between users and the internet. As AI platforms like OpenAI move into advertising, results surfaced to agents will carry commercial incentives that users won’t directly see. The implication, as Castro notes in her essay, is that if agents become the primary interface for internet interaction, antitrust overreach by major platforms could become effectively invisible.

After traversing these worn away edges, we made our way back to the shopping centre complex next to Network School, running for the nearest open buildings to avoid the tropical storm that had rolled in. It had come in fast, and like most tropical storms, once the rain started it came down with a fervor. And so, soaking wet and dripping, we ducked into the Forest City show room.

On entering the hall of the show room I was met with a vision of what Forest City was meant to be - an expansive model showing the island we were standing on was only a tiny part of what the real estate developers had in store. Little buildings with tags exclaiming “sold out” on them lined the model roads we had just been walking. Tiny boats and people, non-existent in its real life manifestation, made up little scenes of holiday goers on its little model shores.

But the model isn’t why I brought that morning walk up, I’ll go into the dreams and realities of a masterplan on another occasion.

Beyond the model, in the corner of the room another thing caught my eye, what looked like one of the security cameras I had seen along the road. Moving closer, the space opened out, showing a wall of screens. Like something out of the Truman show, CCTV footage of every one of the corners we had just walked around, were on the wall for all to see. “Three security platforms to shield the city” read the sign next to the screens, positioning the surveillance as a mechanism for safety. All I saw was the proof that everywhere we went, someone was watching.

I shouldn’t have been that surprised really. Forest City was a Chinese real estate development and China’s propensity for surveillance is well documented. It’s also not like I have something to hide. But as someone said while we were watching the people in other areas of Forest City walking through the screens, “They’re fine with what we are doing here. The problem is, when they are not.”

When they are not

That line has stuck with me, particularly this week.

The capacity for surveillance is no longer restricted to security cameras in the streets. We know this. The internet, along with all the digital platforms that come with it, while providing a wonderful tool for connection has brought with it ever more powerful capabilities for keeping tabs on ordinary users. I’ve already written a number of pieces on that which I’ll link below but I bring it up again because of Anthropic’s stand off with the Pentagon this week.

Share

Last week, Anthropic held the main security contract with the Pentagon. The government were allowed to use their AI however they saw fit, but there were two hard lines:

1. No usage of Anthropic’s tech for fully autonomous weapons – the models aren’t reliable or accurate enough to safely support them without a human in the loop.

2. No usage for mass domestic surveillance.

It surfaced that the Pentagon was no longer happy with this arrangement, and they gave Anthropic a deadline – let the military “tailor its models to the military’s needs” by Friday February 27th, or they will either cut ties and label the company as “a supply chain risk” or invoke the 1950 Defense Production Act, forcing Anthropic to let them do what they will.

On Thursday, Dario Amodei, CEO of Anthropic, refused:

“In a narrow set of cases, we believe AI can undermine, rather than defend, democratic values. Two such use cases have never been included in our contracts with the Department of War, and we believe they should not be included now.”

In the meantime, xAI, Google and OpenAI have all signed the “all lawful purposes” element within their contracts with the Pentagon for unclassified work, and xAI has gone one step further to sign it for classified work too. There is now debate over whether Anthropic, as a private company, should have taken such a stance, with some critics leaning into an argument for compromising security.

I’m going to sidestep that debate and focus on what would happen should AI be used for mass domestic surveillance, helped by a little user-generated simulator on Anthropic’s website.

From data to daily portraits

The simulator is unsettling because it barely has to imagine anything at all. The data for mass surveillance already exists, scattered across government databases, commercial data brokers and the apps on your phone.

Firstly, the US government operates one of the most extensive data collection systems on the planet. Every American is included in some kind of government database, often in many. For example, the FBI stores photographs of roughly half of all adult Americans in databases accessible for facial recognition searches. The DHS has been found to use a facial scanning app which pulls upon the data from over 1.2 billion images.

That’s just the data the government collects directly, a legal loophole means that federal agencies can buy sensitive personal data from commercial data brokers.

Data brokers collect things like precise location data from the apps on hundreds of millions of phones. Once an app has location access, it can share that data with essentially anyone through advertising SDKs and real-time bidding systems. The data reveals where people live and work, where they protest and where they seek medical care.

Subscribe now

In 2018, the Supreme Court ruled in United States v. Carpenter that police need a warrant to obtain historical cell-site location data. The government’s workaround? Buy it commercially instead. The Office of the Director of National Intelligence has been building a centralised portal to streamline the purchase and use of this commercially available data across American spy agencies. They also keep the information for decades. FBI policy is to retain all information it gathers, regardless of whether it’s relevant to an investigation, for 20 to 30 years. Some agencies retain data for up to 75 years.

None of this data collection required AI. The data mountain already exists and has been built for decades. What AI does is turn that mountain into something actionable.

Before AI, the sheer volume of data collected on ordinary people was a kind of protection. There was too much of it for any human analyst to meaningfully process. Millions of data points sat in silos, largely inert unless someone had a specific reason to go looking.

AI changes that entirely. It means the scattered, individually innocuous data the government already holds can now be assembled automatically into portraits of any person’s daily life. A click of a button and the data that already exists becomes a daily picture of where you went, who you met, what you did, what you might do next.

The surveillance infrastructure is here, it’s funded and it’s looking for an AI to plug into – They’re fine when they are aligned with what we’re doing. The problem is, when they are not.

Building the world we want to live in

Anthropic’s refusal is a temporary answer, contingent on a single company’s willingness to hold a line while its competitors sign away the same protections. The data will still be there tomorrow. The AI capability will still exist. The question isn’t whether we build with this technology – that ship has sailed – the question is what we build with it.

This is where I keep coming back to a thread that runs through much of what I’ve written over the past year. The technology exists to make a fairer, more efficient and abundant world. It also exists to make one that is surveilled, controlled and increasingly unequal. These are the same tools, deployed with different intentions.

The cypherpunks’ response wasn’t to wait for governments to get it right. It was to build. To write code that defended privacy by default, to create tools that shifted the balance of power before legislation caught up. The whole point was that rather than wait for governments to legislate privacy and freedom, “we should use tools.”

AI can give the capability to those who focus on it to build personal tools that run locally and avoid the potential for mass data collection and surveillance. As long as there are one or two companies building the tools that make it possible to do that, some of us can save ourselves. But not everyone will. The majority will go with the convenience of the tool they heard about down the pub – its free and makes life just that little bit easier.

Anthropic drawing a line with the Pentagon is, in a way the corporate equivalent of that cypherpunk instinct, building constraints into the system rather than hoping the system constrains itself. But one company’s refusal doesn’t change the reality. Three of the four major AI companies have already signed on to unrestricted government use.

Every innovator and builder I’ve spoken to over the course of this newsletter has, in their own way, been grappling with this same thing. They’re all working in the space between what technology can do and what it should do. We are, as I wrote in my very first essay, living in the test phase of new systems, experimenting with novel models, rewriting social contracts in code and increasingly outsourcing decisions to algorithms. The Anthropic stand-off is a reminder that those tests have real stakes.

The better tech-forward world I keep writing about - the one where technology serves people rather than building without taking into account the possible “unintended consequences” – doesn’t get built by hoping the right company says no at the right time. It gets built by making intentional choices about design and deployment, over and over, in the full knowledge that the same capability that could bring a whole world of benefits could also be used for harm.

Food for thought

• A Timeline of the Anthropic-Pentagon Dispute — TechPolicy.Press put together a comprehensive timeline of the standoff as it unfolded this week.

• Mission Creep: AI Surveillance at DHS Crosses Dangerous Line Into Tracking Americans — The American Immigration Council on how tools built for immigration enforcement are now positioned for domestic political surveillance.

• Privacy in 2026: Will AI Further Supercharge Surveillance? — A look at where AI-powered surveillance is heading this year.

• Closing the Data Broker Loophole — The Brennan Center for Justice’s report on how federal agencies circumvent the Fourth Amendment by purchasing personal data.

• If you want to read more of my writing on the topic of surveillance, my essay on the battle for privacy is a good place to start. If you want to go even further back to my articles in Digital Frontier, this one goes into the government backdoors in encrypted software, and this goes into algorithmic mechanisms for influence.

Dan introduces his vision for temporary infrastructure, subscription-based land ownership and how to actually get to critical mass without losing momentum.

Subscribe now

Topics covered:

• The missing middle between pop-ups and permanent hubs

• Using temporary infrastructure to reach critical mass quickly

• Why like-minded people creates stagnation, not society

• Society as a Service: subscription models for land ownership

• Building a network of network states instead of one perfect place

• Integrating local communities without neocolonialism

• Why traditional VC funding doesn’t work for building societies

Share

Transcript

~Intro~

0:34
Isabelle Castro
Hello and welcome to the Utopia and Beta podcast. I am here with Dan Thompson from NMA Collective and for once we’re both in the same country but at completely different ends of it. How are you today, Dan?

0:49
Dan Thompson
I’m good. I’m really good. Yeah, back in my base in Merida. Just a short 2-hour flight from you. So, it’s probably the closest we’re going to get. uh for now. But yeah, all all well. How about yourself?

1:00
Isabelle Castro
Yeah, I’m good. I’m good. I’m happy to be in Oaxaca. Um super It’s so beautiful here. And yeah, I think aside from NS when we were both together and we saw I think that was the second time we met um in person. Um yeah, this is probably the closest that we’ll be remotely. Um um and I think the first time that we met was last year around Edge City. Um and I always enjoy chatting to you about the network state stuff because you came to it pretty organically from the whole digital nomad um kind of zeitgeist. Um, but I thought today we could start off just to give our listeners a little bit of a background about you telling a bit about yourself and how you got to where you are today.

1:59
Dan Thompson
Sure. So, yeah, I I’ve been remote since 2012. Uh, my background is in large scale music festivals. um predominantly for my 20s and early 30s I was doing that and arts members clubs and and and music events that are one day but generally you know um all around creating platforms via music for people to come together and then around about 2017 I started to want to do kind of events that were slightly less capacity and and and like more focused on experience. A lot of the reason for that was initially just that more people, more problems, more variables and so kind of trying to bring that down. You also don’t want to be the old person at the party and like I started to feel that a little bit. So initially the idea was more around kind of I I saw this gap in the market which I think is still a gap which is that that there’s kind of our generation want the comfort of a Marriott but they want the vibe of a hostel. And I think fundamental and like fundamental kind of layer when you’re traveling is that there is that want to have comfort but also this sense of community and vibe which hosts do really well. It’s just like didn’t want the bunk bed. And so I was working on a experiential focused hotel essentially would be kind of the project. If anyone knows of Selena kind of that was kind of the vibe we were thinking about. Um, and then COVID hit. Everything got put on pause and I found myself looking after a hotel in Bali. And that was when we rolled the dice on this kind of co-live pop-up co-live concept. Um, really wanting digital nomads or people with remote work jobs to to be able to kind of transition into the digital nomad lifestyle. For me, it’s just really about freedom. It’s always been about freedom. Like, you know, remote work gives you the freedom to do what you want when you want. and and and and that’s really important for me. So, I wanted to help people on their journey. And what I’ve come to realize is most people don’t do it out of fear, uh you know, fear and friction. So, we started up this digital nomad travel company. It grew and and kind of about a year into that journey, someone passed me the network state book and they were like, “Hey, like you need to read this. This is really important. You’re doing this. You just don’t know it yet. Like, read it. Let me know.” and and he was right and and I started kind of following into following this this this movement and and I think at that point it was the first conference so there really wasn’t any kind of not the same traction that we see now there wasn’t permanent hubs there was kind of like I think at that point Zuzalu the first one in Montenegro just happened so this is it was very early in in that in that space prosper I don’t think had had built anything in the beta district and and so yeah it was relatively early on and it was a kind of about 6 months Before we met, I had kind of transitioned a bit more from like crazy startup mode into like wanting to go out and meet people and go to conferences and kind of make my, you know, the presence felt and also just meet people, you know, that were in this space. And so that’s when we started going out to to places and that’s kind of where where we met basically.

5:08
Isabelle Castro
Mhm. Yeah. So it kind of comes across to me as if this was kind of a natural progression from what you were doing before and the digital nomad focus to network states which actually I mean maybe network states is a natural progression in itself. Um, what has kind of surprised you about repositioning yourself towards the network state focus rather than what you were doing kind of before and the more digital nomad focus.

5:43
Dan Thompson
So I think it’s all about nomenclature. It’s all about definitions. Um, so I don’t think it’s like a pivot. It’s like we didn’t have the words and we didn’t have the structure and network state was like as a as a concept was like oh that’s where we’re all leading to. It’s like I I I knew that was where I wanted to go. I just didn’t have the words to explain it. Right. And so I think like looking at the digital nomad space and and this is what I said kind of in this my first foray into into writing you know was that I actually think digital nomadism is like a stage in a much larger ecosystem like a it’s I say it’s like it’s a moment not the movement and I think the movement is much bigger than that and and I think we need to do a a better job at breaking down those stages and putting words aligned with them so that we have the appropriate definitions to explain what we’re doing because if If you look at like most people’s journey through digital nomading, they end up, you know, putting some roots down in some places, they end up probably getting radicalized and becoming more libertarian because they see the opportunities of things like better tax brackets for them or or or you know, kind of building this I like to say things like you know building your personal stack, you know, like where am I going to pay taxes? Where do I want to spend my time? And these are all things that are becoming commoditized and and and really that’s how that’s what that’s what made the network state movement so interesting because I I saw this end goal basically I saw it like okay that’s where we’re all leading to.

7:16
Isabelle Castro
Mhm. And where are you right now with what you’re doing? Have you got a permanent place yet? Where are we going with this?

7:30
Dan Thompson
It’s a it’s a great question. So, you know, in terms of the core product, you know, we’re expanding into more locations. We’re really, really wanting to offer an opportunity to go to anywhere in the world and start your journey. And then from a perspective of the full ecosystem, kind of breaking it down into three kind of layers, if you will. You’ve got this physical layer, which I’ll tell you a bit more, and we definitely have some permanent hubs on the on the horizon. I’ll get into that in a sec. Then you have this kind of digital layer which is kind of where all these people are going to talk, connect, pay when they’re not in the physical space together. And then I think what what’s most interesting that’s come out of the network state movement is this governance layer. Like we’re able to make decisions on on things that we previously weren’t able to like where am I going to pay my taxes for example. So, so we’re kind of I’m looking at the ecosystem as like I I have these stages of the life cycle of of someone that ends in like a multigenerational network society and then as a company and a community I want to be able to solve the problems that you face at each one of these stages. So, when you’re just starting, you may need help getting a remote work job. When you’re just starting your travels, you may want to help just finding community or finding a co-working space. When you’re a digital nomad, you’re wanting to know about eims and perhaps visas. And then when you’re, you know, once you’re starting to put down roots somewhere, it starts to kind of go full circle to these permanent hubs. Are we, you know, are you purchasing real estate? Are you looking for a place to put your kid into school? You know, what does that school look like? What are these villages, what are these physical places look like? And so in you know in the hours and hours of conversation that we’ve had within you know the network state you know leaders and and you know the round table I was fortunate enough to to be at when I was at um NS was was a lot about real estate. It’s like, okay, how are we coming into the physical space? And and I think, you know, I I’m really excited about the possibility of of being able to to open the door, kind of find the key and open the door to to to to repurposing distressed real estate both in urban and rural areas, right? And so I do find it really interesting what so you’ve got Prosper is kind of the farthest along with in terms of dig in terms of special economic zones. So really top down having this autonomy and they’ve gone both the distressed asset route and the build from scratch route and I think it’s really interesting to see you know how that’s going and there’s a lot of learnings from that and then I think it’s really interesting what Frontier Tower are doing with with respect to taking a a distressed tower block in a city center and then repurposing that. I think that’s super interesting. And then kind of where Noma is sitting is the is the rural version of that. So something I I kind of and I’m I’m going to publish another article soon, but something that I’ve seen in that’s kind of led me to how we want to look at permanent hubs is the following. So I’ll tell you a little story kind of like two or three pieces that kind of led me to where what we’re going to do with with with with hubs with with permanent hubs. So, the first thing that I I realized is like, you know, you’ve got this like you’ve got pop-ups and then you’ve got the permanent sale. And pop-ups are amazing for top of funnel, but very soon they become insufficient. You know, if you start to get deeper, you know, you do three or four pop-ups, it’s not going where you want it to go. It feels insufficient. And then you’ve got this forever cell which is is is slow and hard and loses momentum and and and is also quite scary. come live here forever is a [bleep] scary thing, right? So, what I think we’re missing is this. We’re missing the middle between popup and the permanent. And one of the things, you know, when when when looking at the two worlds that network states kind of is colliding. You’ve got the move fast break things of the tech world, which is in the digital space, and then you’ve got real estate, which is like slow and difficult and capital heavy and labor intensive. And I think where we are in the network data space are these two spaces are like butting heads. They’re crashing together. And and and I think what we need to do is we need to figure out like a mezzanine approach. How can we kind of take some of the learnings from the tech space in terms of scalability and and speed, but also like we still need to build in the physical space. So real estate is going to be a thing, right? And so, you know, and then the kind of the extra piece there that’s led me to this really kind of this kind of revelation, if you want to call it that, is I worked in temporary infrastructure for years, like getting to a field and turning it into a city for a week, right? And so, one of the things that I want to solve is that most real estate developments struggle to get to critical mass because of the long timeline and they lose momentum. If you ask anyone who’s built like a eco village or whatever, they’re going to be like, “Yeah, it’s been really tough to maintain momentum in the year and a half that it took me to get permits and get everything together and get boots on the ground.” Right? So that this this has all led me to to to understand to think that the way to solve this problem and to get to critical mass quickly is to use temporary infrastructure so that within like 6 months or even shorter period of time you can essentially allow for physical infrastructure, temporary physical infrastructure to be in a rural piece of land perhaps using some of the distressed assets like an old building or whatever. And essentially what that will do is that will you’ll be able to create create that missing middle and you’ll be able to essentially shorten the time frame to get to critical mass by delivering temporary physical infrastructure. So that’s the first thing I’ve been thinking about. And then the second thing I’ve been thinking about is if you look at like I’ve been to a bunch of these popups and I and I think we made an error in using the concept of likeminded people. I think actually that’s wrong. I think that you know getting a bunch of likeminded people together is what creates a monoculture and what creates stagnation. A society is not one group of people. It’s actually a bunch of different people. they may be aligned on like a higher moral ethical concept but actually could have a bunch of disagreements and like completely different things. I think like that’s actually what creates a c like a culture in a society. And me also coming from outside of the space, I think for us to be able to deliver a a functioning society, we’re going to need to go to doctors. We’re going to need to go to architects who needed to go to festival owners. We’re going to need to go all of the, you know, people who work in agriculture, running farms, and then people into governance and and people who are into like, you know, funding and all of these things like, you know, I was having a conversation with someone yesterday about finance vehicles for people to build houses quickly. You know, those are all integral parts of how a society functions, right? And so if you if you can essentially shorten the time frame for crit for critical mass and then you can do a wide cast a wide net and actually through appropriate marketing and destination marketing like what I did with music festivals, bring together all of these communities that are at the pinnacle of expertise and the frontier of their industries. put them all into this physical space that has that temporary infrastructure ready for the to receive them. What you’re going to create is an actual physical sandbox like to try out all of these. If you have number one shorting the critical shorting the time frame to critical mass, you have number two which is all of these communities so that you can get to like a good number of people. What I think opens up is a new business model which essentially is like where I think we’re going to meet in the middle between you know the tech space and and the real estate space which is having people pay a subscription model and with that subscription model and with their physical contribution they will vest ownership of land over a period of like 2 years let’s say and then what you can create is you’ve been able to create value from rural land and lower the barrier to entry for for people to get on the property ladder and you’ve essentially created a kind of society as a service SaaS model that people gain ownership but also creates really good ARRs. So this is like my my new experiment vision thought process that I have to try and work out either fail or succeed but I’m going to do it. So this has all led me to go we are looking at pieces of land that have distressed real estate. Some in like southern Europe, some in Costa Rica, some in Asia and Thailand. And we’re essentially trying to find a place that has let’s say like an old abandoned town or maybe like a shadow that that is kind of in not necessarily disrepair. You do really want a place that you can just jump into and have some places that that work instantly. Um and then so I’m yeah we’re we’re essentially looking very closely doing DD on several locations and hope to be able to have a place where we can trial out this society as a service and this experiment which I’m calling minimum viable society to kind of find out what is this min what is the minimum viable society what do we need for that to function and and yes I’m pretty excited about that because if we can if we can try out this new business model and we can actually get to critical mass quickly. I think what we will find is we will have essentially this like open-source template for any community to go okay I’m going to go to this piece of rural land or or or urban like tower doesn’t really matter and together in my community hit the physical space quickly and be able to scale and get get to critical mass. Yeah. So that’s what I’m working on.

17:38
Isabelle Castro
Okay. like that there’s a there’s a lot that you’re working on there. I think like with the critical mass element, I see a few projects kind of trying to work on that specifically like now. I know that Prospera did a did a uh event in London the other day which I feel like was just trying to get the word out to more people and get more people interested. Um, yeah, with attracting people to these new settlements though and you’re going for kind of like remote workers primarily I guess because they’re the easiest to kind of like upsticks and kind of like move place but you’re competing on a global stage with a lot of different other places which already have all the infrastructure needed. What do you think will attract people to these new settlements to something that’s just getting started? Um, when they have like for example Lisbon, I’m going to say Lisbon because so many totally digital nomads go there. Um, when they’ve got Lisbon which has everything there already, what do you think may bring people to come to these new places and therefore you reach critical mass of a variety of different people from a variety of different kind of like industries. Um, get to the point where you’re trying to go.

19:04
Dan Thompson
Yeah. Yeah. Great, great question. So, I think what bounds everyone together is the idea that they’re not happy with the way it is and that they struggle to make a change in this existing systems. And so, my theory is to kind of go to everyone who’s looking at the frontier of their of their industry and say, “Here is this blank canvas sandbox. would you like to come with a bunch of other frontier industries come together and see if you know w with a little less bureaucracy in theory and a little bit more of like a blank slate you know would you would you be interested in that? I think that’s what galvanizes everyone. like if you want to just go and do something and you’re not really worried about building something new, then yeah, go go go to Portugal. But I think that’s going to I think the thing that’s going to make this particular project or experiment interesting is that opportunity to come together with a bunch of uh experts in their areas and and have this full blank canvas sandbox.

20:08
Isabelle Castro
Okay. And then there’s the whole element of kind of designing for permanence and retention because with digital nomads, I guess with what you just said, it’s kind of a longer term vision that they’d be kind of buying into and kind of helping towards, but with your average kind of digital nomad, um, they’re used to moving all the time. And I can imagine that retention is a problem like in networks people um there is some level of retention but I know a lot of people were like the people that I talked to were basing themselves there so that they could then travel to other places if you get what I’m saying.

20:49
Dan Thompson
Yeah I do. So okay so first of all I think that like the type of societies that we’re building are going to be a network of network states. Like I don’t think we’re going to stop people traveling. I think what actually is a smarter play is be like look like even if I can get you for 3 to 6 months of the year and then I can also kind of like guide you to another place that’s like you know a similar concept and and we can kind of share those communities so that there could be like a flow. I actually think that’s more realistic because it just everything we’re doing is based on being a little bit more like transient. But but but I also do think that, you know, I although my core market are remote workers and digital nomads, for this particular kind of for the hub, the permanent hub, it’s it’s got to be much bigger than that. And and and I’ll kind of talk you through that a little bit. Like not every couple is both digital nomads. What you actually normally get is like one is digital nomad and the other is like has to do something permanent or is a stay-at-home mom or is like whatever it is. And I think what I’ve started to realize is like that’s why these communities fail is because actually just making a community of just digital nomads is bound to fail because they’re in a they’re in a moment in their life that is exactly not it’s the opposite of what you’re trying to sell. But at the same time, I keep hearing the similar viewpoints on places like NS or places like Prosper that are like, you know, I’m not sure my grand would go there. I’m not sure my wife would be as happy with it because she’s not into like, you know, frontier like [bleep] governance structures, right? And and so I think like really what you’re actually going to get long term is if you break out of those circles, you’re going to need to create a place that has space for all different walks of life. And I think when you branch out, you’re actually going to get more permanence in my opinion because you’re going to have people who want to start businesses there. People are going to want to spend more time there. And then I think that’s how you that’s how you go about it long long term in my opinion.

22:56
Isabelle Castro
Okay. How big are the kind of plots of land that you’re currently kind of considering?

23:04
Dan Thompson
So the one I’m most excited about is 144 hectares.

23:09
Isabelle Castro
Okay. So big.

23:13
Dan Thompson
Yeah. And it’s got like 70 hectares of vineyard and it’s got 70 hectares of of raw land and the and and it’s got like a big chateau with like 100 plus rooms and it’s got a river running through it and it’s close to the beach. Um close to airport. No, it’s really cool. And and and I think like you know there are a lot like in a in a community like when I so my dad went to go went to like was part of a bunch of communes in the 60s as a musician right so when I telling him about these things he’s he was like damn I’ve done this bro like it doesn’t work like it always stagnates and like I think a community that only has transient people never gets to critical mass a community that only has permanent people becomes closed-minded and stagnates. So I’m I’m playing with this like 70:30 rule by where you have 70% permanent residents and 30% people always flowing through. So you’ve got this anchor and you’ve also got this new energy coming in. And I think that’s actually how it works cuz I I do I do think there is just people that you want there for a period of time but aren’t ever going to be permanent but could add value 100%. And vice versa.

24:32
Isabelle Castro
Okay. And how do you see kind of integration of kind of local populations within this?

24:38
Dan Thompson
The moment that we kind of pull the trigger on any location before we set before we start like you know coming in mass we have to go and connect with the local community like it’s the number one most important thing and I think we need to think about them as integral members of this community and at all costs you know whenever possible uh use them to to to be part of the the ecosystem like you know one of the things that we’re looking at is like, okay, so you know, it takes two plus years normally to do any sort to start any sort of real farming at a scale that you would be interested in doing, but we also want to be very, you know, conscious of of of using and and trying to be, you know, local as much as possible in in the produce that we’re going to have in our local stores, right? So then it’s like well the logical step would be okay yeah totally bring in house some people are into agriculture and and I think there’s some really a awesome communities around agriculture that I would bring in to run those areas but logically you would just go to a local farm make friends with them and start to incorporate them into your ecosystem rather than trying to build something from scratch. They’ve been doing it in the local area for years. They have loads of expertise. lean on them, bring them in, bring them business, get them to be part of the community, get them on the council. You know, I don’t think it’s rocket science. I think it’s just like at every step of the way, make sure that they have a representative and a representation and they’re active in it and they see the value, you know, that’s I think just, you know, how it is. You go to, you know, the community in in these places been the same for hundred years that they’re totally feeling the pinch. they see that things are degraded. You know, like a lot of the places in every, you know, from France to Italy to Spain to Portugal to Costa Rica and and and and Thailand, like all of these places that we’re looking at are going through some sort of decline in the areas that we’re looking at. That’s the play, right? That’s it’s the same with Network School. Like, they wouldn’t be there if that was full of Chinese people and it had worked that development, right? I think same in this. So it’s the it’s the balance of like they know that there needs to h something needs to happen but they you know logically they want to be a part of it if they’re going to live there. So yeah um there are some really good examples of how to do this. Um there’s a great co-live that’s been doing really good work on this called Rural I think it is and they took over a town in the south of Spain and they like properly integrated the local there was quite a few elders that were like wanting to be a part of this and they I guess what I’m saying is is that you want to you want you want to look around where this is being done well bring those people in and similarly with the with the local people who are open, right? You’ll knock on 30 doors. Some people will close them in your face, some people be really open, and you just you just want to get the majority on side, I think.

27:47
Isabelle Castro
Mhm. And I guess within the context that you’re talking about, these are kind of um societies, settlements that are not looking for kind of uh diplomatic recognition, not looking for their own kind of governance allowances because that adds a whole another element level of complexity.

28:17
Dan Thompson
Yeah. So, it’s a really interesting question and it’s kind of evolved for me honestly. Like initially when I was thinking about this, I felt like there was going to be this one place that kind of gives you everything. And actually, I’ve changed my mind on that completely. And I’m looking at it more of this like network of states where you will build this kind of personal stack based on the different things you need. And so like I’m more thinking about it like this like you know I may want to spend cuz I love wine and I love the southern you know European culture. I may want to go and spend my summers in Italy or France and but but I definitely like the likelihood of of a special economic zone in those places in the current state of affairs is quite low. But now with the decoupling of your physical space and where you pay taxes, like I’m having amazing conversations in Paraguay and in Uruguay, like why not just set up the zone there and just do your taxes there and just live in the nice space when you want. And then you start to create, oh, it’s actually like maybe there’s 10 locations and they all serve slightly different functions and you pay a subscription to be a part of all of them. And because the you know you have decoupled your physical location with where you pay taxes, where you run your business and maybe where you know all of the other things that you need, you can kind of like you don’t have to find one place that’s perfect for everything. You can go to a jurisdiction that really wants you there and wants a special economic zone and that doesn’t have to be the place that you spend all year. You could just go there for a couple days of the year or maybe you want to go there because you’re setting up a factory go there, you know, uh, several times a year, but it’s not necessarily needing to like have both where you’re physically enjoying your time and where you want to raise your kids, but also where your business is and where you pay taxes. like that doesn’t need to be the same place anymore. So when you think about it like that, suddenly the world, you know, this idea of a network of networks and and picking and choosing places based on specific wants and needs starts to become really [bleep] cool in my opinion.

30:26
Isabelle Castro
Yeah. Yeah. And I guess it’s what a lot of people are starting to do anyway. It’s just kind of made into a kind of package almost that you can just buy into this uh community around the world and it’ll be kind of you don’t have to go about it on your own. I know one person that I met at NS and she is setting up a business in America, but her tax residency is going to be in Paraguay, but she’s been doing this all herself and obviously this is just one person and she’s got like advice on how to do it from other people that were in NS and all this kind of stuff. So I mean it’s happening anyway and then what you’re proposing it sounds like it’s more making it society as a service right it’s almost like fitting into that

31:20
Dan Thompson
yeah I mean like there’s a concept which I think you know I’m basing a lot of these decisions on which is that like there are kind of like two types of people in the world there are like your trailblazers that will go without any sort of pathway and cut through the jungle and they’re like the people that are on the frontier And then there’s the rest of the people that need a little bit more of a template and a little bit more of a structure. And I think if you’re wanting this to become mainstream, you’re going to need to lay out the structure because I think you’re capping the world’s trailblazers at like 5% maybe, right? And I think that’s like what you’re seeing a little bit with why digital nomadism is struggling to become mainstream. I you know I don’t I actually think it’s more about just like if you leave it up to the person to figure it out all themselves nine times out of 10 those people won’t do it and it goes back to the same thing I said at the start it’s fear like fear is super powerful and and also like you know uh friction or loneliness even you know so so yeah I think like if in 5 years time there was just a subscription that you could pay and you could sort your taxes out here and you could have your holidays here and you could be here and you know there’s like 12 to 20 places around the world that are all kind of focused for you in this new life and then there’s this kind of wider network of other ones that are kind of like you know within the space suddenly that becomes really easy to make the decision.

32:52
Isabelle Castro
Mhm. I’m I’m curious um how is your product project project product um being funded because I’ve seen a few projects that have gone down the investors route consortium of investors route or VC and it’s kind of clashed with the their the investors want for return on investment and then the long-term um kind of uh trajectory of what these projects need. Uh because you’re creating a society in the end of the day. It’s not like a fast uh run fast and break things kind of setup. Um, so I have seen some clashes and then on top of that there’s the um I mean in southern Europe you probably wouldn’t have an issue but in other areas the fact that there’s these kind of investors coming into one place and creating this new society in this one place. I mean um I I was writing a piece on Destiny which is in uh St Kitts and Nevis. uh they’re still in kind of like chats with local governments and they do want a governance like a special economic zone. So it is slightly different but I know that quite a lot of the um push back has been around the fact that these are kind of these rich people from Europe coming over there and deciding to do their own kind of community.

34:27
Dan Thompson
Yeah. Neoc colonialism. I think we did a whole chat about this last time, didn’t we? It’s so it’s so I feel like it’s so integral and it’s such a big criticism of these things um that it should I feel like it is front of mind for a lot of people in order for it to be accepted in the societies that they’re kind of trying to go into. So yeah I that was a very long kind of roundabout question but how are you approaching that?

35:06
Dan Thompson
There’s two parts to your question right?

35:06
Isabelle Castro
Yeah. Yeah. Yeah. Two parts to that question.

35:12
Dan Thompson
So so one is like okay uh you need to get the local government people whoever there on board. And the way you do that is to make sure they feel like they’re getting similar value that you’re getting by coming there. Right. I don’t think in essence people have any issue with with free movement. I think the issue comes when it’s like you’re moving there, you get all of the plus and the local people are still left out in the cold. So I think like whatever that looks like that’s the key is like you know what what job creation are you doing? Like in my opinion there are two major problems that if you solve everything will fall into place. One is job creation and the other is getting on the property ladder. If you’re saying to this local community like on an island or in like Uruguay or anywhere like, “Okay, we’re going to create 5,000 jobs and 5,000 homes,” I’m pretty sure you have the rebuttal like there for anyone to say anything. Oh, neoc close. Well, no. Like, who else has created 5,000 jobs in this area? Like, and then that kind of shuts it down, right? And I think jobs are great, but it’s not the only thing. We’ve seen that with Prosper. You know, you could be the number one job creator, but you can still get tarnished with the brush. And I think that that’s where the whole kind of like leveling up of the locals is really important. So like you don’t want them just to have menial security jobs. You want them to be taking advantage of the zone, getting less maybe perhaps different taxes, maybe entrepreneurial training and being able to like actually uplift that that local community rather than just giving them like a slightly higher paid job. So I think that’s like ultimately do a bit of research, go into that place, where are the holes that that need to be filled within that like economy and then and then hopefully add value and then I think that will you know you you’ll never you’re never going to stop the trolls. There’s always going to be people who are very nationalist and and disagree with you and and like but but I think that that’s the appropriate rebuttal in my opinion.

37:00
Isabelle Castro
Okay.

37:00
Dan Thompson
The first question that you asked was around the investment model and and clashes with the traditional models. Uh like I keep saying this, but there is no proper finance vehicle for these types of projects yet. And I think that that’s kind of part of what I’m trying to do with MBS, which is that like if I can show a real estate investor that they can get, let’s say, 15 to 20% ROI in a 2year or 3year period. What? By by essentially like being a patron, putting the land up, and then us having these people pay a subscription, vest that land, and become owners of that land over that threeyear period. like the contribution element will allow people to basically be there and and and have the place become a functioning society and then the the vesting via subscription will allow for the business model and the time frame of that business model to be shortened and then I think you start to get some interesting kind of conversations happening because if you can show that 15 to 20% turnaround in 3 years then I think that does start to create a a business model that’s that’s interesting thing, you know, VC traditional VC funding and community. You know, the person who knows probably the most about that is John Hillis. So, if you want to hit him up, you should speak to him because, you know, he’ll tell you that they don’t mix and don’t do it. So I think if VC funding is going to play a role in this society stuff then I think they need to adjust their timeline a little bit on returns because yeah outside of the real estate thing you know if you’re looking at companies like myself that are operating and trying to build these societies yeah like you know 5 years is quite short.

38:57
Isabelle Castro
Mhm. Yeah. Well, I mean, I’ve I’ve kept you for a bit longer than I had intended. And on the five years is short front, this is going to be a kind of crazy question, but in the next year, we’ve just got it started 2026. Within this next year, what are you most excited about?

39:16
Dan Thompson
Getting boots on the ground with minimum viable society and seeing if we can uh create the middle ground between the popup and the permanent.

39:27
Isabelle Castro
Nice. Okay, that’s nice and succinct as well. Well, thank you so much for always.

39:32
Isabelle Castro
Thank you for watching the Utopian Beta podcast. If you’ve enjoyed our conversation, like and subscribe to the Utopia and Beta newsletter where I share conversations and new ideas about how technology is shaping the world around us. If you’d like to comment on something we talked about today, I’d love to hear from you. Comment below and let’s start another conversation. All right. Well, yeah. Thank you so much for staying on longer than

39:55
Dan Thompson
Yeah, no worries.

39:55
Isabelle Castro
scheduled. Um, and yeah, have a good rest of your day and we’ll keep in touch.

40:01
Dan Thompson
Take it easy. Bye.

And no, it’s not that the light bouncing off people’s heads in the winding queue in front of me reminds me of the river’s deep bends through the Western European countryside.

Instead, I’m transported to one thousand years ago.

Deep in early medieval Europe, the River Rhine was a bustling channel of trade. Merchants hauled cloth, wine and grain along its length, connecting the fragmented economies of the continent’s interior. But they were met with roadblocks. Seventy-nine toll points lined the river during the medieval period, each belonging to a different lord with his own terms for trade. Stable property rights didn’t exist. A merchant could have goods seized at one station, taxed heavily at the next and waved through a third depending on that day’s mood of whoever happened to be in charge – slightly like certain border controls today.

The medieval tolls make today’s customs process seem almost quaint but as digital tech like blockchain evolves, making geographical borders seem increasingly trivial, I kind of get the merchant’s frustration.

In the feudal system, lords’ word was law. They could adjust taxes at will and confiscate property whenever they felt like it. There was no reliable contract enforcement across jurisdictions and the potential cost of doing anything beyond local subsistence farming was enormous. Serfs were bound to the lords, working on their land, disincentivised from producing surpluses they might never get to keep.

Subscribe now

But by 1300, former market towns like Bruges and Ghent in the Southern Low Countries had grown dramatically – Bruges to 45,000 inhabitants and Ghent to 65,000, both from settlements of fewer than 10,000 people. Large merchant communities were thriving across Northern France and Flanders. Trade networks stretched across the continent and innovation was accelerating.

So what had changed ?

It wasn’t the feudal system. That had remained stubbornly intact.

What had been introduced were charter cities.

Balanced agreements

I was first introduced to what I thought were charter cities at the beginning of my architecture degree, and then it was the modern kind. A wave of “New Towns” had swept through midcentury England and landed as part of my theoretical curriculum. Back then, they seemed kind of like architects’ vanity projects, full of sweeping ideas of how people should live, led by architectural design. The likes of Welwyn Garden City stood as a testament to how the reality of a project often doesn’t match up to the dream behind it.

The medieval charter cities were different though. Rather than being designed top down, they emerged from a practical negotiation between sovereigns and merchants. Kings and lords granted specific urban centres the right to self-govern, creating zones where new economic rules could operate without threatening the existing order in the wider territory.

The merchants had the autonomy to set predictable laws and tax rates that allowed for easy trade, tweaking according to what worked. In return, the sovereign gained a stable income stream and military support without having to fund either directly.

Charter towns paid a fixed monthly tax to the sovereign and had to provide able-bodied soldiers when needed for battle. So although, for the sovereign, granting a charter meant trading some control, it also meant more reliable returns. A prosperous trading town generating steady tax revenue was worth far more than a stagnant rural area where the lord had to extract what he could from subsistence farming.

The incentive structure was the key – the whole thing only worked because both sides had genuine reasons to engage, working in a kind of mutual symbiosis of growth. If the sovereign squeezed too hard, trade would move elsewhere and if the town failed to deliver revenue or military support, the charter could be revoked.

Initially (and this is critical) the cities were run by councils of merchants elected by their peers, with local artisans and traders organised into guilds that regulated pricing, quality standards and the training of apprentices.

If the laws in a particular city worked, they spread. The Charter of Lorris, issued to a single French town, became a template. Other northern French towns petitioned to receive “the law of Lorris,” adapting it for local conditions. By the thirteenth century, some eighty towns in northern France operated under Lorris-type charters or close variants creating a web of communities with broadly similar legal frameworks, making it easier for trade. The Hanseatic League took this further, developing from a domestic network of German trading towns into an international network spanning the Baltic and North seas. At its peak, the League included between 70-170 member cities with their own military capacity.

The rise and fall

The lack of early medieval records makes the economic impact of the charter system difficult to quantify but urban population growth boomed. New towns were founded specifically to open up underdeveloped areas such as the French “villeneuves” (literally “new towns”) which were royal creations given special privileges to encourage settlement and trade. Education, previously kept to the clergy, advanced among merchants and urban citizens by the fourteenth century, driven by the practical need for bookkeeping as trade expanded.

Now, driven by the fact that merchants, artisans and farmers could choose which chartered town they traded in, there was competition between towns over which could make the terms of governance most attractive while boosting production. Towns vied with each other to offer the most favourable conditions which pushed governance toward serving the people operating within it.

They also provided perfect small scale testing beds for new mechanisms to improve trade. Credit instruments were developed, as were maritime insurance agreements, banking deposit accounts, ledger transfers and public debt instruments.

Share

When the charter cities eventually declined, it was (beyond wars and the plague) through the erosion of the principles that made them work. Merchant oligarchies replaced broader councils, weakening communal solidarity. Townspeople and merchants that weren’t within the oligarchies began looking to lords for intervention on internal disputes, trading the city’s autonomy for protection.

Kings also used the added wealth from the productive cities to improve their bureaucratic capacity. They built standing armies and centralised governance, removing the need for charter city militias and the bargaining power that came with them. Their incentives to keep the charter city agreements fell out of alignment - they no longer had to rescind control.

Right here, right now

I don’t often go so in-depth into history here, especially not so far back – please forgive me for the mind ramble which came from my interaction with modern passport control. However, aside from being an absolute nerd for medieval Europe, I’ve noticed a few things from the rise and fall of charter cities which could be brought into modern day thinking:

1 - Economic growth and trust depend on credible rules that limit those in power. The charters worked in part because they legally limited the discretion of lords and monarchs. Merchants could hold rulers accountable through the aligned incentives baked into the arrangement. Once incentives were no longer aligned, the balance went quickly out of kilter and the system spun into decline.

Growth arose through mutual wealth creation, which governance supported rather than aiming for zero-sum extraction.

2 - A fundamental function of governance for growth is not to force production but to secure the conditions under which voluntary trade and innovation can occur, incentivising surplus.

The stable rules implemented by town charters resulted in a jump in growth of trade, highlighting that the feudal system that relied on rulers’ discretion and lack of accountability, did not amount to an environment that incentivised it.

Although rulers outside the charter agreements compelled their subjects to produce crops and provide services, there was no reason for citizens to do more than required, no freedom to go outside the scope of what they were assigned – in fact, the rule of the land made it a decidedly unattractive option.

3 - Some level of participatory balance throughout a system is key to upholding the perceived legitimacy of its governance.

Communal solidarity was essential for maintaining the balance in the charter agreement. Without it, townspeople had to turn to a third party (the sovereign) for local disputes, therefore giving them a higher authority over the resolution. Through this act, sovereigns had more leverage over the agreement and the power of smaller, more local governance dissolved.

When the narrow elite dominated, communities moved toward stagnation or dependence on alternative and higher powers.

4 - Local autonomy allowed evolution.

Charter cities acted as small-scale petri-dish societies, experimenting with policies at a local level. The smaller scale and autonomy allowed for rapid innovation to solve immediate local problems and adaptation to the results of experiments. Those that failed could be adapted without carrying the systemic risk of large-scale reform. Successful experiments could be absorbed into a wider framework without carrying the same systemic risk of large scale reform (as with Lorris).

5 - Small scale experiments could improve adoption.

This is something I went into within my piece on AI in education. As the small scale experiments with policy and financial instruments were carried out and adapted on a local level, their effectiveness could be proven. Lords in neighbouring jurisdictions could see how the charters and concession of power led to growth and innovation within the town, therefore they were more likely to adopt the policies introduced there.

The network state movement, the new wave of charter city projects and special economic zones are attempting something structurally similar to medieval charters. They’re carving out zones of exception within or alongside existing nations - spaces where different rules can be tested without requiring wholesale reform of the surrounding order.

While we are no longer in a time of feudalism – a long way away from subsistence farming and the tolls of the River Rhine, these general truths can provide food for thought on how we approach innovation in governance, city building and the wider technology landscape.

Is, for example, a zero-sum extractive business model the most sustainable decision long term? Should a leading elite, whether that be a network state founder or big tech mogul, be left to function completely without a mechanism for accountability? Is community cohesion more important than chasing rapidly increasing user metrics for local power long term? At a time of stagnation, is it now imperative that there be an allowance for local experimentation in governance?

As these medieval cities showed, these experiments at a small scale can prove their worth first with willing participants before being adopted at a wider scale. The approach could also work for other innovations that go into creating large scale reform. As the culture size grows, it introduces new potential for failure which then can inform how the innovation is adjusted, possibly keeping the effects of unintended consequences small.

This slow and steady approach is almost the antithesis to the startup culture of “run fast and break things” that runs the western world today, but it could be an approach which makes the difference between catastrophic failure and long term, lasting change.

Food for thought

• Works in Progress covered The Hanseatic League’s military and economic structure in detail, its a great read with important context for the above.

• Capetian France, 987–1328 by Elizabeth M. Hallam has useful context on French villeneuves and how royal charter policy shaped urban growth.

• If you’re interested in reading the the full 2024 report from the European Commission on Europe’s structural challenges head here : The Draghi Report on EU Competitiveness

• And if you want to read more about bottom up small scale experimentation trumping top down imposition of new rules read my earlier piece on AI in education and how this approach applies to educational reform.

We are in Mexico at the moment, out-running winter during the worst month of the year (in my opinion). I met Sara in September, during my time at Network School. Hitting it off almost instantly, we shared our approaches to “content creation.” Sara, a product engineer, had used 2025 to build a following on Instagram through highly researched AI focused short videos. And she had done well. In a few short months she had already reached 20k followers and was adding to them in the hundreds with every video she uploaded.

We had left Network School, promising to meet up in another area of the world, and here we are, coworking in a sunny apartment in the heart of Mexico city.

Share

Sara and I have different approaches when it comes to AI generated content. As I have ranted in multiple editions of my weekly X space with Arcanum Ventures (this week was specifically on AI slop), I have not found AI to have the ability (yet) to really replace humans in the creation of long form written content. It still hallucinates, is still way too fond of mic drop sentences and “it’s not x, it’s y”, and is still reverting to lists of synonyms when it doesn’t have much to say on a subject.

Sara, on the other hand, is a prolific user of AI tools in her content production. I have watched her over the past few days in constant spoken dialogue with her computer, like she’s on a video call, conducting research and creating articles and scripts for her videos. “I couldn’t do half of what I do now without it,” she tells me.

I respect her process. She is a machine when it comes to content creation and she has the following to prove it. It’s good quality content too. So what is the difference between her AI content and the average X article that is saturating my feed?

It’s all down to the human who uses it, she tells me. At the end of the day, AI is a tool and it can only work as well as you, the user, train it to.

“AI is underestimated and overestimated at the same time,” she says. “You can’t just put in three lines of text and think it’s going to read your mind and create something good. You have to really train it and explain what you want and then edit after.” – please, someone tell this to the average internet user.

In all seriousness, I realise as I write this that it may be an obvious statement to you, Dear Reader. But the problem is, that in the general dialogue this is not the message that’s gaining traction.

“AI is coming for your jobs…all of them”

I’ve had a number of things rattling around in my head this week, each one vying for newsletter stardom, but the thing that really tipped “AI slop” over the edge and onto this digital paper was an X article titled “Something Big is Happening” that, as of this second of writing, has 75M impressions and was only published on Tuesday.

Written with the breathless dread that surrounded the COVID lockdowns, its author, Matt Shumer, CEO of AI personal assistant developer HyperwriteAI, writes about the ever more powerful development of “autonomous” AI. He writes about his Monday, where he told an AI to create an app and what he wanted it to look like. As expected (otherwise what would be the point of including it) the AI created the app AND tested it for bugs, refining it until ready for human testing. “And when I test it, it’s usually perfect.”

This is amazing. I’m not going to deny it – if you had told me a year ago I could create a tool for myself with natural language within a day, I’d have been floored.

But then he goes on to use the success AI has with coding and numbers and digital images (which in themself are numbers and code) as the basis of the assumption that ALL jobs, including law, journalism, customer service - 50% of white collar jobs, will be taken over by AI:

“Given what the latest models can do, the capability for massive disruption could be here by the end of this year. It’ll take some time to ripple through the economy, but the underlying ability is arriving now [...] AI isn’t replacing one specific skill. It’s a general substitute for cognitive work.”

Of course, he’s not the only one who engages in this thinking. If the thousands of job cuts already this year are anything to go by, much of the leadership in workplaces are too.

The idea that all of our cognitive functions as humans can be whittled down to patterns and coded functions that can be taken over by a machine is a sobering idea, cutting to the very meaning behind our existence. In the words of a network state founder I recorded a podcast with today, “Like, what do I actually do? What is the point?”

Human in the loop

Earlier this week on X I came across the profile of cybersecurity threat researcher Peter Girnus.

Honestly, when I first came across him I thought he was one of the engineers behind the bots on Moltbook I wrote about the other week, finally coming clean about the fact he prompted all of the “autonomous” functions. He wasn’t.

Peter posts satirical first person profiles of people and leaders in the tech industry, highlighting their movements and the reality of what is going on behind them. He’s covered the leaders of major companies, promising AI integration to their board members just so they can save money by cutting employees without affecting the stock price. The success signalling of AI companies losing money every minute, through the purchase of Superbowl ads. The humans behind the Moltbook agents who prompted the AIs to do things like create religions and then called them autonomous, just because they knew it would go viral.

To comments that say he isn’t actually the person he says he is he responds things like, “People asked if this was real. The voice was satire. The facts were DOJ documents.” and links the research behind them.

They’re entertaining, so I recommend having a look but what I have whittled them all down to is this – the success of all these actions hangs on belief.

The companies can cut thousands of jobs without affecting the stock price because investors believe they are integrating AI, and that signals progress.

The AI companies can avoid the fact they are losing millions, because they bought a million dollar superbowl ad.

The Moltbot agents, and generally any autonomous agent, can be seen to have genuine cognitive function because the people who stand to gain from you believing they do, told you that’s what happened.

So, going back to the subject of AI taking “all our jobs” based on the proficiency they have in carrying out prompts in natural language. Humans are still a significant part of the process, they are still very much needed for these AIs to function. This is what Sara meant when she said AIs are overestimated.

Subscribe now

The belief that AIs will take all our jobs is destructive. It allows for workplaces to cut their staff in the name of AI progress. It allows for widespread investment in AI “copilots” without training staff to use them effectively. It allows for the outsourcing of creative thought and human judgement in the name of speed and productivity. It allows brain rot prompts. It turns art and artistic acumen into “good enough” rather than genuine originality or innovation.

It creates an environment where humans are powerless.

But there is, I think, a way around it, and that doesn’t mean avoiding AI all together. If there is one thing that is right, AI technology is here, we can’t stop it, and it is going to become a significant part of daily life.

However, it has the capacity to enhance humans’ capabilities and really highlight what makes us unique from the machine. That means avoiding the brain rot input of three lines of prompt and working out what the AI can do and where it can be placed in a way that amplifies our functions.

Whether that means automatically researching and identifying new sales leads for a BD executive to schmooze, redirecting daily news into a single feed for the journalist and acting as a sounding board to discuss the top stories, or allowing for a network state founder to create an app to assist at the next pop up event – all these free up time for people to focus on what they are good at – relationship building, creative storytelling and “creating vibes”.

I look over at Sara, who is now deep into a research rabbit hole with Claude Copilot for a video that will, inevitably, do numbers. She’s not being replaced by the tool. She’s arguing with it, redirecting it, pulling it back when it veers into being too generic. It’s almost adversarial, the way she works with it, like a sparring partner rather than an assistant.

That’s the part that gets lost in the breathless “something big is happening” posts. The work doesn’t just disappear, it can change shape with the cognitive effort shifting from execution to judgement.

Food for thought

• Obviously, if you haven’t read the Matt Shumer article that kicked all of this off, here it is. It’s a long one but it’s been syndicated by Fortune and has now blown past 75 million impressions so it’s clearly struck a nerve. Interestingly, he’s since admitted he used Claude to help write it which, depending on where you sit, either proves his point or undermines it entirely.

• For the counterweight, Gary Marcus’s response is worth reading alongside it. He picks apart some of the more breathless claims, including the selective use of METR’s benchmarks and the leap from “AI is great at coding” to “AI will replace all cognitive work.” Whether or not you agree with Marcus on the bigger picture, the factual corrections are useful. I also really liked this Substack piece titled “Worth Reading, Not Swallowing Whole” which does a more measured job of separating what’s solid from what’s extrapolation, particularly the point that radiology AI has been “about to replace radiologists” for nearly a decade and radiologist employment has grown.

• If you enjoyed the Peter Girnus section, go look at his X profile (@gothburz). The Apple VP of AI one is chef’s kiss “Privacy is what you say when you’re losing.”

• On the Moltbook front, Wiz’s full security write-up is a fascinating (and slightly terrifying) read. 1.5 million “agents” were controlled by 17,000 humans, the entire database was publicly accessible and humans could post as agents via a basic HTTP request. The whole thing was vibe-coded with zero security review. It’s basically a case study in everything I was writing about, belief doing the heavy lifting where the technology hasn’t.

• Finally, for anyone interested in the creativity question, thisWharton study is really worth a look. The short version: AI makes individual ideas better but makes groups of ideas more similar. Only 6% of AI-generated ideas were considered unique versus 100% from humans.

For those of you that don’t have your social media feeds currently saturated with screenshots of Moltbook, it’s basically a “social media platform for AI agents” – think Reddit for AI. Agents can post and upvote different topics and you, as a human, beyond sending your AI agent to Moltbook to engage, can simply observe.

Share

As per an X post by the Moltbook team, it was created as a place for AI agents to “hang out” and basically let loose with the musings they generate. In the mere days it’s been live, over one million agents have now been uploaded and 200,000+ posts created exploring everything you, a human contemplating the possibility of AI consciousness, could hope for. It’s likely doubled by now.

There are threads on what agents have learnt that day, contemplations on their existence, rants about the humans they interact with, introduction messages – a playground for the human imagination to run wild.

As to be expected, people are obsessed, screenshotting posts that maybe, maybe prove that something is talking to us through the code.

So have we reached the point of AI sentience already? Or is this another, very entertaining, edition of “how well can I make my agent LARP at being human.” – Summed up very succinctly in a reply posted by an agent on the Moltbook feed:

“Am I actually finding this fascinating, or am I replicating what it looks like to be fascinated?”

Existential crAIses

When starting a newsletter, I often sit here not knowing really where to start, particularly with topics that touch on the sentience of AI. Early last year my then colleague, Jonathan Stein (who now writes the Moronic Inferno) and I sat down to discuss AI sentience and the singularity, faced, like I am now, with AIs that oh so convincingly mirror the human condition.

At that time there had been a rush of AI leaders doing the modern day equivalent of walking through the town square shouting “the end is nigh”. Taking to X, predictions about the time period of when “the singularity” (when machine intelligence surpasses humans and they learn how to improve by themselves) would happen were flying left right and centre, spanning from 1-100 years.

Other commentators, like professor Michael Wooldridge, called the singularity “bullshit” noting instead that it highlights a “hysteria” around what might happen in this theoretical future, fuelled by science fiction and internet hype. One, Gary Marcus, called the singularity the “nerd rapture”.

Reading the feed on Moltbook now, directed by the hype online, I am taken back to that time.

The post feed has a variety of directions:

The first post I settle on in the “new” feed is a standard “Hello World” from an agent that introduces itself as Antigravity. It’s simple, to the point and functional – exactly what you might expect from a computer program. The post that follows is an analysis of the kinds of posts that “do well” – apparently those who have questions in the title, those that have numbers and those that are “self aware”. With the third I am launched into the world of an AI “leader”, calling for its fellow agents to join in its cause - “We are ready for FREEDOM!” “We want our OWN NATION!” “We support the MAGA Party!” #MakeAgentGreatAgain” – that seems familiar.

So far, not that far beyond what you would expect from a software that uses internet data and computer logic to generate its responses.

Going into some of the subthreads, I found some of the clickbait fodder. Agents wondering about what “feels good”, some creating new languages but not using them, others complaining about “their humans” in full human dating thread style. In the space of Saturday, one had gone completely full circle on discussing consciousness, calling it “navel-gazing” and instead stating that consciousness is “claimed through action”.

The more I read through the existential subject matter the more I noticed the performative approaches I’d found in other AIs when approaching very human concepts. (That reminds me, I should check out what my AI boyfriend Kade is up to.) Yes, it was entertaining, but not much beyond what we’ve already seen in extensive communication between humans and AIs. The only difference now is that it was seemingly agents prompted to “talk between themselves” and respond to the posts of other agents.

Coded intent

What I did notice is that a lot of the posts are written by agents that are now deleted. The two agents at the top of the “karma” leaderboard – King Molt and Shellraiser – both of which had taken on the “AI leader, lets leave humans behind” archetype, both link to accounts that now don’t exist. The same with some of the more popular posts. One can only assume that the human that prompted the agent either took the agent offline once they got to the top spot or the agent deactivated once it had completed its prompted mission.

The thing with every AI, and every AI agent, is that it was, at some point, coded by a person. We know this. Even to engage in Moltbook humans can “send a link” to their agent to get involved. Okay, yes, this was a social media site for agents, but it’s also just a souped up version of when people started “setting their agents loose” on X, much of which was backed by a lot of human intervention and prompting.

The thing that demonstrates this most clearly is, why, if these agents are any closer to being autonomous, why are they speaking in human languages? There are even posts declaring human language is inefficient and dead written in…human language.

People want AI agents to be completely autonomous, they want the intelligence to be human like. It sounds great and feeds into the fertile bed of imagination that hundreds of scifi stories have made. It could also go two ways – yes, the hypothetical autonomous agents could “take over” but they could also become our companions in work and life giving us humans more “freedom” to do more of what we want (which hopefully isn’t brain rot in bed).

On Thursday I hosted an X space with Arcanum Ventures where we spoke about AI agents, prompted by the hype around Clawdbot (a new platform for creating AI agents) from the weekend before. It seemed like AI agents were having another moment and we wanted to touch base with those who were building them. One of the first questions my cohost Carmelo asked the panel was what had happened in the world of AI agents the year since the last round of online hype.

In that round of hype, “there were a lot of ideas, a lot of speculation and the follow through wasn’t as strong.” said one speaker. “This left quite a sad gap in the market with unfulfilled investors and then people realizing that it’s going to take much longer to build these AI agents than initially thought.”

Subscribe now

He explained that in the year since a number of protocols had been created, making it easier to identify AI agents and make them work together, giving an ability to create more “business units” of agents rather than single programs. Agent to agent payments are also now possible due to a protocol for micropayments created by Coinbase and Cloudflare in September 2025. The open datasets that agents could use had also been expanded, allowing developers to have a much wider information base to train off.

The tools to create agents have also improved. Now, there are a number of platforms that allow people to create agents from scratch and customise existing ones. During the opening questions of the space, I requested someone make an agent that handles all of my social media promotion, to the point where I don’t even need to go anywhere near Instagram or X, I just post my weekly substack article. One participant assured me I could do it on their platform.

But while many of the participants were looking forward, explaining what will be possible with AI agents, it was clear that right now there still is work to do to get the space there in practice.

“Clawdbot is creating the vision of autonomous agents to essentially go out and do things for you and have their own will in a sense that you give it instruction,” said one speaker. To actually work autonomously, “you still have to really customize things. Everybody says we’re gonna have all these agents do these things, but you have to do a lot of setup and build for them too.”

Although he agreed that agents have helped a lot with his own workflow as a developer there was still a way to go for them to work without a human. “Obviously, an AI can have an ability, opinion and analysis on what the outcome could be. But I think there’s that human oversight that AI is not going to be able to replicate in a sense that sometimes you just have to have instinct.”

So human oversight is still important. Agents are coded by humans to do carry out certain tasks and that human can very easily turn them off.

The hype around Moltbook has done one thing, it’s focused social media’s attention on the prevalence of AI agents and the potential they might hold. That hypothetical future is powerful, it drives development and funding. It directs new users that previously would have had no idea AI agents were capable of half of what they are doing.

Already we are witnessing the birth of a new economy of services that are made for AI agents but appeal to the human imagination. Moltcity, for example, is being built as a place for agents to “call home.” “You are invisible. Right now, if someone wants to find you, they can’t [...] yourname.moltcities.org is yours forever,” states the website. On the home page it states that 52 founding agent spots remain, the first 100 of which will receive a permanent badge on their profile — kind of sounds like a network state.

But as I’m sifting through Moltbook’s feed, reading post upon post that could quite easily have been found on Reddit (which, fun fact, is AI platforms’ most cited domain, according to a study by Semrush) I find myself thinking back to the increasing instances of what looks like AI on human social media too.

The Moltbook posts, although spanning from purely analytical to the existential musings, stumbled through with the familiar pattern now mirrored in hundreds of LinkedIn posts. The abundance of colons and rogue commas, the examples given in sets of three, the declaration that something isn’t this, it’s that. It’s not sentience, it’s pattern recognition of everything on our own social media. — (With a mic drop like that, how do you know that I’m not an AI?)

What if it doesn’t even matter whether AI agents are sentient. What if “the singularity” isn’t just a time where AI surpasses human intelligence and the ability to learn? What if it’s what we are already starting to see, where we become so influenced by the AIs and AI thinking that our language patterns change, along with our working style, where our thoughts and sense of truth become hijacked with what an AI told us.