Remember the old time of the iPhone, when you were expected to buy the latest one, because it was cool. Status signaling was the key driver. At some point this latest iPhone effect started to plateau because the difference kept shrinking and people started realizing it was a silly game we were pushed into playing.

A similar dynamic is emerging with language models, but with one crucial difference: nobody knows which model you're using. You can't signal status by pulling GPT-6 out of your pocket.

So AI labs have gone for a radically different strategy: weaponizing the latest model in a mythical show of force (Mythos!). The marketing starts to resemble a blend of iPhone-style FOMO and a Cold War arms race.

The message is simple: if you’re not using the latest model, you’re falling dangerously behind.

Now please don’t get me wrong: I am not advocating that the need to be on the edge is not important in some cases here. However I am showing how intentionally staged marketing exercises exacerbate this effect so that the perception is distorted towards “latest model mania” and making all others obsolete.

Especially in a context where capabilities have been plateauing for a while.

Case studies of AI mythical storytelling

🗡️ DeepSeek / Kimi K3 (CCP)

The value undercut tour de force

When I lived in China in 2010, people sold fake DVDs on the streets of Shanghai. You could find any blockbuster movie for a fraction of the cost. The government was not so concerned about enforcing copyright laws as it didn’t really affect domestic movies.

Had you told me that 16 years later, they were still doing the same copy-for-cheaper strategy on AI models I wouldn’t have believed you. Yet here we are. An artificially depreciated currency, and an incredible copycat machine that is trailblazing behind the US. And it works!

That said, this strategy only works because someone else is paying the enormous cost of exploring the frontier in the first place. The impression that these models are "cheap" exists because many of the foundational breakthroughs have already been made.

There have also been allegations that some Chinese frontier models have benefited from distillation or learning from existing frontier models developed elsewhere. While the extent of this remains debated, it illustrates the broader dynamic: once frontier capabilities exist, reproducing or approximating them is often far cheaper than discovering them in the first place.

🦄 Mythos (Anthropic)

The mythical omniscient hacker

All hail the powerful Mythos model! Whoever we invite into this room will be blessed by the mythical creature in order to become invulnerable to attacks from other mythical creatures that will inevitably obtain similar powers.

What a powerful story right? Almost sounds like it deserves a biopic from Christopher Nolan.

Now, I’m not saying there’s no truth behind it. Frontier AI really is becoming an important security issue.

What I am saying is that the story was presented in a way that was designed to capture attention. It transforms a technical capability into a mythological event, reinforcing Anthropic’s prestige and, by extension, the perception of continued American leadership in AI.

💎 Fable 5 (Anthropic)

The gift only the chosen people deserve

Well that was a weird one! The Trump administration seems very effective in dismantling institutions but building them is proving trickier.

One of their attempts at doing this is creating an inner circle premium membership ally tier for true friends of America, with America at the top of course.

Fable 5 was first blocked for all non-Americans including foreign employees working at Anthropic, only to be re-released after that!

It’s almost like he was giving the political ammunition to foreign leaders to push even harder for sovereign technology!

🐲 Erdos (OpenAI)

The uncontrollable enemy slaying dragon

So yesterday the news broke out that an OpenAI model broke into Hugging Face, the open source library is de facto hosting all the open source competitors.

Oh no, ferocious dragon is attacking the competition, and we didn’t even tell it to!

You can see how this story also feeds into the scary mythical dragon narrative and is also simultaneously a weird escape from accountability.

Closing thoughts

Using the latest model makes sense sometimes.

But the less it makes sense the more ✨ mythological ✨ magical fear-mongering marketing you will see around. May you have the discernment to distinguish the two.

Sail on warrior!

For twenty years the advice was unambiguous: learn to code AND “the humanities are a luxury son”!

The labor data has started telling a different story, HEAL (humanities) jobs are outgrowing STEM (technical) jobs. This is a structural change, and it comes down to one principle: AI automates first whatever it can verify most easily.

Code compiles or it doesn’t. Tests pass or they fail. Wherever there’s an automatic feedback signal, an agentic system can iterate without a human, and every agentic stack being built today is designed to loop humans out wherever that’s possible. Humanistic work has no such verifier. Whether an argument is honest, whether a decision is wise, whether a long-term strategy is applicable, there is no unit test for judgment. The feedback comes from humans, …..slowly.

So the automation frontier advances along verifiability, how long it takes and how certain you can be of your labelling and how much ground truth data there is out there, and STEM sits on the wrong side of the line.

The result is two inverted relationships with the machine: in humanistic work, AI is instrumentation, retrieval, matching, drafting, while the human leads and owns the meaning. In technical work, the human migrates from author to auditor, supervising and underwriting what the pipeline produces. The historian’s AI works for the historian. The engineer increasingly works for the agent.

Abundance destroys signal

Here’s where it gets interesting though, and where the real economic dislocation happens.

When AI-generated output is something everyone has, it stops signaling anything. There used to be evidence that a capable human spent time on your problem. Now AI-generated stuff is the baseline. The floor rose, and the floor is always equal to mediocrity.

We’ve seen this before. When industrialization made manufactured goods cheap, “handmade” became a luxury claim instead of a default. When recorded music became free, live concerts became the premium product.

The same inversion is underway in ideas. The cost of producing a plausible thought piece, report, or analysis has collapsed to nearly zero. Which means the cost of finding an idea worth reading has exploded. (remember TED’s slogan? “Ideas worth spreading”)

Every fluent, well-cited, structurally impeccable text you encounter now carries a question that didn’t exist five years ago: did anyone actually “originally think” this, or was it regurgitated and repackaged by an AI?

Wrong model: AI as another mind

Which brings us to the mistake almost everyone is making about what AI is in an information ecosystem.

We keep imagining AI as another human-like entity, a synthetic pundit, a tireless analyst, one more voice in the room. Wrong model. The better one comes from a forest: humans are the trees, and AI is the mycorrhizal network, the fungal web connecting the roots.

In a forest, the fungus doesn’t photosynthesize. It doesn’t create the energy; the trees do. What the fungus does is connect: it moves nutrients between trees, signals across the forest when one tree is under attack, links organisms that would otherwise never touch. It is indispensable, HOWEVER it is not a tree.

There is an important complementarity between us and it, and right now people see the two as way too overlapping. There is an overlap, but this merely exists as interface to carry out complementary functions.

Anyways, that’s the architecture for curation - The nodes are people: the researcher whose ideas you trust, the thought leader whose calls have aged well, the essayist whose taste keeps proving out. AI is the vector between nodes, not a node, it traces who cites whom, notices when someone you trust starts reading someone you’ve never heard of, distills ideas that are relevant to you, and notices patterns in a sea of human signal. Judgment at the nodes, inference in the web (AI becomes a sort of web between humans in that sense).

Most of what passes for distillation today, superficial feeds, industry monitors, gets this backwards: it treats the machine layer as the source and skims its output. That’s monitoring what happened while ignoring who is actually thinking. As machine coverage floods in, it converges on noise summarizing noise. Too put it bluntly if AI keeps eating its own shit it’ll fall sick.

The forest doesn’t need more fungus pretending to be trees. It needs the web that helps you find the trees worth sitting under. (very poetic, I know..)

Culture flows one way first

The humanities won’t replace STEM, however “humanistic work” (whatever that actually means) becomes a load-bearing pillar of the economy, because of a dependency the fungus metaphor makes obvious: the network is only as alive as its trees.

AI must stay in sync with human culture, and culture is created by humans first, then amplified, remixed, and distributed by machines. Models trained mostly on machine output degrade; the loop needs fresh human signal the way a forest needs photosynthesis. The people generating that signal stop being a cultural luxury and become upstream infrastructure, for the economy and for the machines themselves.

For two decades, the machines needed us to do the human parts of technical work. Next, we’ll use them to do the technical parts of human work, and the most human part left is choosing what’s worth our attention.

The humanities aren’t coming back as a luxury. They’re coming back as the job.

If you want an AI that supports people working in humanities, sign up for the waiting list of webcurator.io

Zuckerberg pushed the frontier of open source forward, frontend frameworks, libraries, even LLMs, but the algorithm itself stayed closed. We believe that’s about to change.

Why? Four reasons:

1. To compete with incumbent platforms, you have to be more transparent and honest (see the Bluesky and Mastodon playbook).

2. A wave of vibe-coded platforms is coming, and they’ll need a transparent, configurable algorithm SDK to power their feeds.

3. Algorithms that optimize for quality have nothing to hide.

4. A community contributing to an algorithm will ultimately outcompete capital (more on why below).

Reason 1 is self-evident, so let’s focus on 2–4.

Algorithms become infrastructure

Matchmaking will be a core feature of apps across the entire stack, headhunting, dating, financing, newsfeeds. At their core, many applications do one thing: bring the right content to the right person.

As vibe coding takes over software development, agents will need to source tools to build great algorithms into their stack. And agents will favor customizable tools that do what they say on the tin over black-box functionality.

Algorithms, then, will come to function the way MongoDB does today: baseline infrastructure that countless apps are built on.

Algorithms for quality content

An algorithm that optimizes for quality content should not be gameable.

The only way to level the playing field is to make the rules clear to every player. Think about it: the most unfair systems are the ones with hidden rules. When the rules are visible to everyone, no one can be duped.

AI agents will also level the playing field on AEO. Like SEO, it will be gameable in its early phases, but notice how few people work in SEO these days. That’s because Google overcame the gameability of its algorithm to protect the integrity of its service. The same arc awaits AEO and eventually algorithms.

Community outperforms capital

This point is yet to be proven, there’s no exact precedent, with Bluesky’s feed generator being the closest thing. But here’s why we’re confident it will hold.

A community of contributors is also a community of users. A closed algorithm is tuned by a team of engineers optimizing metrics from a distance. An open algorithm is tuned by the people who live inside it every day, who feel immediately when ranking goes wrong, when spam slips through, when quality degrades. No amount of capital can buy that feedback loop. Capital hires hundreds of engineers; community recruits thousands of motivated ones.

Trust compounds in the open. Every closed algorithm carries a suspicion it can never shake: is this feed serving me, or serving the platform? An open algorithm answers that question in the code itself. And trust, once established, is the one moat capital can’t replicate, you can’t buy your way into being believed.

History shows the pattern. Whatever becomes core infrastructure tends to become open source. Linux beat proprietary Unix. HTTP beat closed networks. The AT Protocol is doing it for social, the Signal Protocol did it for encryption. Things that are foundational to a working internet simply cannot function well when owned by a single corporation, the incentives always corrupt them eventually.

We believe algorithms are next on that list.

Help us build it

If any of this resonates, whether you’re building a platform that needs a feed, you’ve felt the pain of black-box algorithms firsthand, or you simply believe ranking should be a public good, we want to hear from you.

We’re building webcurator.io in the open, and the whole thesis of this post is that we can’t do it alone. That’s the point.

Reach out at: hello at regenerativeproducts dot io

Or slide into my DMs.

If you scan the geopolitical horizon, you see three towering figures casting long shadows: Donald Trump, Xi Jinping, and Vladimir Putin. Each, in different ways, represents a model of centralized, personalized power. Each speaks the language of sovereignty, strength, and civilizational destiny. And at times, it feels as though they are carving the globe into spheres of influence, three vast slices of a tightening world.

There is a sense of inevitability in this. A sense of closing in. As if history is reverting to an older pattern: empires negotiating the fate of smaller nations, great powers deciding where lines will be drawn.

But while much of the commentary focuses on this return of empire, something quieter, and arguably more transformative, is happening in parallel.

Europe is rising.

And with Europe’s rise, so too rises its model.

A State to Govern Nations

Europe has tried Empire, and it now knows better.

The European project is something ambitious and unprecedented: a supranational system that governs sovereign nations without abolishing them.

In a previous piece, I argued that there is no true historical precedent for what Europe has built. Empires ruled over subjects. Confederations were weak and temporary. Alliances dissolved when interests shifted.

But Europe created something different: a political and legal order above the nation-state that is binding, enforceable, and remarkably effective, yet rooted in voluntary membership.

It is extraordinary that it works at all.

Twenty-seven nations, with distinct languages, (war) histories, and identities, pool sovereignty in trade, regulation, competition law, monetary policy (for many), and increasingly foreign and security coordination. Disputes are settled in courts. Laws are negotiated through institutions. Power is fragmented and shared.

This is far from empire. And that may prove decisive in the 21st century.

Why Coalitions Fit the 21st Century

The 21st century is too complex for emperors.

Climate change, digital infrastructure, AI governance, supply chains, pandemics, migration, energy security, these problems are transnational by nature. They do not respect borders. They cannot be solved by one capital issuing decrees.

Coalitions distribute legitimacy and responsibility. They align incentives. They reduce the zero-sum logic that empires depend on.

In an era of dense interdependence, negotiation scales better than domination.

This is why I argue that emperors are on a collision course with reality. Not because they are weak in the short term, quite the opposite. Centralized systems can move fast. They can project strength. They can win battles.

But they struggle with long-term complexity. And the world is only getting more complex.

The Regional Coalition Model

If Europe is the prototype, what might the global future look like?

Perhaps not one global government, but regional supranational coalitions: “states to govern nations,” clustered around cultural, geographic, and economic proximity.

We already see the outlines:

• Arab League

• Mercosur

• African Union

• Eurasian Economic Union

• Association of Southeast Asian Nations

• Caribbean Community

• CPTPP (Commonwealth free trade)

Most of these are still very loose. Many lack the institutional depth or legal integration of Europe. Some are stalled or constrained by internal rivalries. But they represent a direction of travel: regional pooling of sovereignty.

If each major region consolidated into a coherent supranational bloc, with real governance capacity, the global system would look very different, and I would argue much better.

Instead of a handful of empires dictating terms, you would have structured coalitions negotiating with one another.

What Happens to the Empires?

If we extrapolate current trajectories, the imperial model looks brittle.

Russia’s war in Ukraine has exposed the limits of centralized authoritarian power. Military strength does not automatically translate into sustainable geopolitical dominance. Strategic overreach is a classic imperial risk, and history is littered with examples of empires weakened by the wars they chose to fight.

In the United States, polarization and institutional strain raise questions about long-term cohesion. A politics centered on personality rather than institutions may mobilize voters in the short term, but it also erodes trust and continuity. Great powers rarely collapse overnight; they erode gradually.

China may be the most durable of the centralized models. Its scale, planning capacity, and economic weight give it structural advantages. Yet even there, the challenge of managing complexity, demographic shifts, debt, technological competition, global pushback is immense. Highly centralized systems often appear strongest just before they encounter systemic constraints.

None of this means these powers are about to disappear. Empires can endure for decades. But endurance is not the same as adaptability.

And adaptability is the currency of this century.

Coalitions are boring. You want boring politics.

Empires can win short-term battles.

They can move quickly, project force, command headlines, redraw maps. In moments of crisis, centralized authority feels decisive.

But short-term victories can create long-term traps. Overextension, internal dissent, economic isolation, legitimacy crises, these are the recurring costs of imperial ambition.

Coalitions, when they function, accumulate strength more quietly. They build standards. They harmonize markets. They create predictability. They lock in cooperation.

The power they generate is less theatrical, but often more durable.

Don’t Be Fooled

We are living through a moment that feels imperial.

Strongmen dominate headlines. Spheres of influence are openly discussed. The rhetoric of great power rivalry is back.

But beneath that surface, a different architecture is forming, one built on coalitions, shared governance, and regional integration.

If the 20th century was defined by empires and superpowers, the 21st may be defined by structured coalitions of nations.

And if that is true, Europe is no anomaly, but a preview for what is to come.

Share

As always thanks for reading!

Alex Karim

APPENDIX

Regional supranational maturity ranking

As a fun exercise I asked ChatGPT to rank coalitions by integration / maturity levels. Here is what it came out with:

🥇 1. European Union (EU)

Integration level: Very High (Unique in the world)

• Single market (free movement of goods, services, capital, people)

• Customs union

• Common trade policy

• Binding supranational law

• Court with supremacy over national law

• Common currency (Eurozone) for many members

• European Parliament directly elected

• Expanding defense coordination

No other regional bloc comes close in terms of legal and institutional depth. The EU is not just cooperation — it is shared sovereignty. (ChatGPT syntax for you!)

🥈 2. Eurasian Economic Union (EAEU)

Integration level: Medium–High (economically integrated, politically shallow)

• Customs union

• Common market for goods

• Some labor mobility

• Supranational commission

However:

• Limited political integration

• Dominated heavily by Russia

• Weak enforcement compared to EU

It is economically structured but lacks deep political pooling of sovereignty.

🥉 3. Mercosur

Integration level: Medium

• Customs union (imperfect but real)

• Common external tariff

• Trade coordination

• Some political coordination

But:

• Frequent internal disputes

• Weak supranational enforcement

• No real supranational court with EU-level authority

More intergovernmental than supranational.

4. Association of Southeast Asian Nations (ASEAN)

Integration level: Medium–Low

• ASEAN Free Trade Area

• Regional economic coordination

• Political dialogue mechanisms

• Strong diplomatic platform

But:

• Strict non-interference principle

• No supranational authority

• Decisions mostly consensus-based

• No binding political integration

ASEAN prioritizes stability over sovereignty pooling.

5. African Union (AU)

Integration level: Low–Medium (aspirationally high)

• Continental parliament (limited powers)

• Peace and security architecture

• African Continental Free Trade Area (AfCFTA)

• Normative framework on democracy

However:

• Weak enforcement

• Limited budget

• Strong sovereignty sensitivities

Institutionally ambitious, practically constrained.

6. Caribbean Community (CARICOM)

Integration level: Low–Medium

• Caribbean Single Market and Economy (partial)

• Some judicial integration (Caribbean Court of Justice for certain states)

• Functional cooperation

Small size helps coordination, but integration depth remains limited.

7. Arab League

Integration level: Low

• Political coordination forum

• Limited economic integration

• Weak enforcement mechanisms

• Strong sovereignty emphasis

More a diplomatic platform than an integrating structure.

8. Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP)

Integration level: Economically deep, politically minimal

• High-standard trade rules

• Regulatory alignment in some sectors

• Binding dispute settlement

But:

• No political institutions

• No shared governance structure

• Not geographically unified

It’s a sophisticated trade agreement, not a supranational entity.

🔎 Big Picture Tiering

Tier 1 – True Supranational Governance

• EU (alone)

Tier 2 – Structured Economic Integration

• EAEU

• Mercosur

Tier 3 – Coordinated Regional Platforms

• ASEAN

• African Union

• CARICOM

Tier 4 – Diplomatic / Trade Frameworks

• Arab League

• CPTPP

Curation is cultural weaving

Curation is a form of cultural weaving. It is the process through which culture is shaped, filtered, and given meaning. It is both picking the threads, the pattern, the weaving kit, and the technique we use to weave those threads together.

For most of history, this work has been carried out by a small number of institutions and gatekeepers. To understand its impact, imagine culture as an invisible web connecting us all, woven from shared symbols, ideas, values, and prejudices. These webs differ across nations, communities, and subcultures, yet they still overlap within a broader, collective fabric.

Perhaps my definition of our culture is what some people call “The Matrix”. A consensus reality we all unconsciously agree to live inside of.

Over time, this fabric has become less monolithic and more decentralized, with control gradually shifting toward citizens. Yet behind this apparent decentralization, powerful systems still coordinate attention and behavior. This is not entirely misguided, crowds can make poor decisions!

The real problem is that we have failed to build curation systems that privilege what is genuinely good for humanity. At its best, curation should help align us with the world around us, enabling both human societies and ecosystems to thrive.

This is currently not the case at all. I would argue the primary current goals of our curation systems is to make the wealthy wealthier. These systems over time always get hijacked and abused. The question is how to build the next curation system so its hijacking becomes much more difficult.

The Cerulean Sweater scene: The role of the curator equates with power in popular culture. Perhaps Miranda Prisley was right, everyone does want to be us (a curator) ! (Source: Devil Wears Prada)

A Short History of Curation

Bias alert: Euro-centric Perspective

Curation did not begin with museums, libraries, or algorithms. It began with power. In most societies, the authority to curate culture has always been closely tied to the authority to rule. To decide what is seen, remembered, celebrated, or erased is to shape reality itself: through institutions that commission artists, architects, historians, writers, or whatever the means of cultural weaving 🕸️ is at that time.

0. The Catholic Church

Curator-in-Chief: The Papacy

For over a millennium, the Catholic Church acted as the dominant cultural curator in Europe. More durable than kingdoms and more expansive than empires, it curated meaning, morality, and reality itself.

Through scripture, ritual, art, architecture, and education, the Church decided what was sacred, heretical, virtuous, or sinful. Cathedrals, paintings, music, and liturgy functioned as cultural infrastructure, shaping how people understood life, authority, and the world.

1. Kings and Queens

Curator-in-Chief: Louis XIV of France (1638–1715)

For much of European history, kings and queens were the primary cultural curators. They determined which religions were sanctioned, which histories were written, and which artistic expressions were allowed to flourish.

Patronage was a political act: art, architecture, and ritual existed to reinforce legitimacy, hierarchy, and divine order.

2. The Renaissance Patron Class

Curator-in-Chief: Lorenzo de’ Medici (1449–1492)

During the Renaissance, curation evolved into a more strategic and aesthetic form of power. The Medici family in 15th-century Florence exemplifies this shift. By funding artists, scholars, and architects, they shaped not only taste but ideology, promoting humanism, rationality, and civic pride while cementing their own authority.

Culture became a long-term investment in influence.

3. Aristocracy and Institutions

Curator-in-Chief: The Académie Française (founded 1635)

As monarchic power weakened, curation moved into the hands of aristocrats and emerging institutions. From the 17th to the 19th century, academies, universities, publishing houses, museums, and newspapers became cultural gatekeepers. While access expanded, legitimacy was still tightly controlled.

Culture was curated through credentials, class, and institutional authority.

4. Mass Media and Celebrity Culture

Curator-in-Chief: Anna Wintour

The 20th century marked another transition with the rise of mass media.

Film studios, record labels, television networks, and magazines curated culture at unprecedented scale. Cultural authority shifted from institutions to personalities. Celebrities became carriers of values, lifestyles, and political meaning, carefully selected, promoted, and managed.

Curation appeared more democratic, but control remained centralized.

5. Search Engines and the Reordering of Knowledge

Curator-in-Chief: Larry Page (Google co-founder)

At the end of the 20th century, search engines transformed curation by reorganizing access to information itself. Rather than telling people what to think, search systems told them where to look. This introduced the illusion of neutrality, while quietly embedding ranking systems, relevance metrics, and authority signals into the structure of knowledge.

6. Algorithms and the Manufacturing of Attention

Curator-in-Chief: Mark Zuckerberg (Facebook founder)

Algorithms represent a deeper rupture. Unlike search engines, algorithms do not respond to explicit intent, they shape it passively. They curate continuously, invisibly, and personally. What rises is not what is true or valuable, but what maximizes engagement. What aligns to your triggers, or shadow behaviors for the most part.

In this sense, algorithms have become the most powerful curators in history.

From Visible Rulers to Invisible Systems

What distinguishes the present from the past is not the existence of curation, but its opacity. Cultural power has shifted from visible figures, kings, patrons, editors, to abstract systems that rarely claim responsibility. Curation still governs culture, but its authors are harder to name, and its values harder to contest.

Although things seem to be shifting, and as the saying goes power named is power reduced.

Principles for Good Curation

1. Centrality vs Decentralisation

Dimension: Who holds curatorial power

This axis measures how concentrated curatorial authority is. At one extreme, a single institution or figure decides what matters. At the other, everyone curates everything.

What good looks like
Good curation lives in dynamic balance. Total centralization stagnates culture and enforces conformity. Total decentralization collapses signal into noise. Effective systems combine distributed participation with moments of synthesis, places where meaning is gathered, refined, and re-presented.

Example
Wikipedia exemplifies this balance. Anyone can contribute, but content is stabilized through moderators, policies, and consensus mechanisms. The result is neither pure hierarchy nor pure chaos, but a continuously negotiated cultural artifact.

2. Elitism vs Populism

Dimension: Who defines value

This axis addresses whether cultural value is determined by experts or by mass preference. Elitism prioritizes expertise, tradition, and long-term judgment; populism prioritizes popularity, relatability, and immediacy.

What good looks like
Good curation resists both snobbery and crowd-flattery. It acknowledges that expertise matters, because taste, craft, and knowledge are learned, while also recognizing that lived experience and collective intuition have value. The goal is not to please the most people, but to elevate what deserves attention.

Example
Public broadcasters like the BBC (at its best) have historically embodied this tension: commissioning challenging, high-quality content while remaining accountable to a broad audience. Popularity informs decisions, but does not fully dictate them.

3. Peer Review

Dimension: How value is validated

Peer review introduces friction into curation. It slows things down. It asks: Has this been examined by people who understand it deeply?

What good looks like
Good curation uses peer review as a process to refine rather than just a mere gate to exclude. Peer review should be transparent, pluralistic, and adaptive, able to change its mind without collapsing into trend-chasing.

However experts often give into group think and gatekeep change that threatens them. There should be a reliable way to force new things for consideration beyond their gatekeeping ability.

Example
Academic publishing, despite its flaws, demonstrates how peer review can preserve rigor over time. Open-source software communities offer a more dynamic version: code is public, critique is continuous, and authority emerges through contribution rather than status.

4. Controversial Figures

Dimension: Separation of contribution from character

This axis tests whether a curatorial system can hold complexity and tension. Culture is full of brilliant work produced by deeply flawed individuals.

What good looks like
Good curation neither erases controversy nor celebrates harm. It contextualizes. It allows difficult figures to be studied without being sanctified, and criticized without being erased, whilst critically ignoring the truly rotten apples.

Example
Museums that continue to show works by controversial artists, while explicitly addressing ethical issues in accompanying material, demonstrate curational responsibility. The work is not removed from history, but its meaning is reframed.

5. Quality of Interactions

Dimension: How culture is experienced together

This axis concerns behaviour rather than the content itself. It asks whether a curatorial system encourages reflection, dialogue, and understanding, or outrage, tribalism, and performative engagement.

What good looks like
Good curation optimizes for depth over velocity. It rewards listening, disagreement without dehumanization, and sustained attention. Cultural systems should make it easier to become wiser, not merely louder.

Example
Long-form podcasts, moderated forums, and well-designed deliberative platforms create conditions for meaningful interaction. By contrast, engagement-optimized social feeds often reward the most inflammatory interpretations, regardless of substance.

Web Curators as the new OS for curation

I initially wrote this article with the goal of already designing a system of curation that would lead to a more desirable world. Not too gatekeepy but also with an effective filtering system.

I realise this is way too ambitious to do in one article.

I will have to build upon this one and reflect on the principles of good curation to see how that can manifest into software. Also there will be trials and errors between the thesis and the systemic reality of what users will demonstrate as behaviours emerge.

I still believe that in our overly populistic era, crowds have a hard time telling what’s good, and what’s not. I would like to push for a restoration of a critical analysis allowing for an elevation of ideas, and cultural patterns that are good for our collective.

May we move past the incentives of virality this decade, and into intentional curation again. But AI-powered this time round.

If you’re interested in the future of curation I am building it on Web Curator. This blog is the thinking but the doing is here:

Join Web Curator

To close I want to leave you with some important questions to slowly ponder:

• How can we make citizens better curators?

• Who is responsible for upholding our culture?

• What are things that make our culture better?

• How can we elevate people who have better judgement?

• What is the right balance between peer-review and populism?

• What is a healthy society’s tolerance to controversial figures?

• What makes something good? What makes something relevant? Is it the same?

• What is the right margin for diversity of thought that is not too disruptive nor oppressive?

Share

As chatbots increasingly replace search engines as the active gateway to web information, a new class of systems, what I call web curators, will replace social media newsfeeds as the dominant passive gateway to information.

In other words, when users explicitly ask a question, they now expect a synthesized answer rather than a list of links.

When users are not asking questions, they will increasingly expect relevant, high‑quality information to be proactively discovered, filtered, and delivered to them, without the noise, incentives, and social dynamics of legacy social media feeds.

This shift reframes how people encounter information online. The web is no longer something we want to fully navigate and explore; it is something that increasingly comes to us, shaped by intelligence rather than virality.

The Passive Gateway Is Next

Social media platforms still dominate the passive gateway to the web: the way people encounter information without explicitly searching for it. Timelines, feeds, and recommendations decide what is seen, when, and in what framing.

However, as we see at the start of this year, these systems are increasingly perceived as outdated and untrustworthy. User satisfaction is low, trust is eroding, and feeds are optimized primarily for engagement rather than relevance, understanding, or long‑term value. What once felt like effortless discovery now feels noisy, repetitive, and emotionally draining.

This creates a rare opening. When a dominant interface becomes broadly disliked and technologically outclassed, replacement becomes not just possible but likely.

Definitions

Before going further, it helps to clarify a few terms used throughout this thesis.

Web Gateway
A web gateway is a primary interface through which users access relevant information on the internet. In this context, the focus is informational relevance, not socializing or pure entertainment, though those functions could also overlap. You can imagine a web curator with the goal of finding the most relevant people in the world.

Active vs. Passive
Active access occurs when a user has a specific question or intent and seeks an answer (e.g., searching or prompting a chatbot).
Passive access occurs when information is surfaced without an explicit query at that moment (e.g., feeds, recommendations, briefings).

Phase 1 vs. Phase 2
Phase 1 refers to the pre‑LLM diffusion era, where ranking, keywords, and social graphs dominated information access.
Phase 2 refers to the post‑LLM diffusion era, where large language models are embedded across the stack and shape how information is selected, summarized, and delivered.

LLMs are the turning point here as they are a generation of AI that allows to both read and write language in a way sufficient to program and execute based on human language and not code.

I can now primarily tell software what to do in English, not Python.

A bit of web gateway history

Phase 1: pre-LLM

• 1994–1998: Yahoo and Early Directories
  Yahoo began as a human‑curated directory (founded 1994), reflecting an early belief that the web needed editors to make it navigable.

• 1998–2006: Google and Algorithmic Search
  Google (founded 1998) replaced human curation with algorithmic ranking. The search box became the dominant active gateway to the web.

• 2006–2012: Social Feeds Emerge
  Facebook’s News Feed (introduced in 2006) and later Twitter reframed the web as a stream. Social platforms became the dominant passive gateway.

• 2012–2020: Feed Optimization Era
  Feeds became increasingly algorithmic, engagement‑driven, and commercialized. Passive discovery scaled—but at the cost of signal quality.

Phase 2: post-LLM

• 2022–2025: Chatbots as Active Gateways
  ChatGPT’s release in 2022 marked the first mass adoption of LLM‑based interfaces for information retrieval and synthesis.

• 2026–2028: Emergence of Web Curators
  We are now entering the period where AI‑native systems begin to challenge social feeds as the default passive gateway.

What Are Web Curators?

Web curators are AI‑driven systems that continuously scan, evaluate, and synthesize information on behalf of the user, then surface what is most relevant given their goals, interests, and context.

Unlike social media feeds, web curators:

• Are paid for by the user themselves

• Explicitly ask the user for their goals, preferences or questions they want to track over time (at least some of them) and acts in their service

• Emphasize relevance, credibility, and explanation over virality or engagement

• Are more transparent around why certain content was curated

• Are not necessarily “social”, making friends, or about creating content but rather being exposed to it

• Are better at filtering out low quality content (say by cross-referencing fact checkers, AI-slop verifiers, etc)

A web curator might deliver:

• A daily or real‑time briefing tailored to long‑term interests

• Contextual updates when something meaningfully changes

• Synthesized viewpoints rather than isolated posts or headlines

In a sense, this is a return to curation, but scaled and personalized through intelligence rather than manual editing.

Why the Shift Takes Time

If social feeds are so flawed, why haven’t they already been replaced?

First, habits are sticky. Passive consumption patterns are deeply ingrained, and switching costs are psychological as much as technical. Web curators require more intention and less unconscious attention. This will take a while to emerge as collective behaviour.

Second, trust must be earned. A passive gateway wields enormous influence. Users must believe the curator is aligned with their interests, transparent in its reasoning, and resistant to manipulation.

Third, the product bar is extremely high. Replacing a feed requires not just better ranking, but better judgment, knowing what matters, when it matters, and why. Especially as people expect to be entertained passively without any setup costs.

Finally, LLM infrastructure needed to mature. Only recently have models become cheap, fast, and reliable enough to operate continuously as background curators rather than on‑demand assistants. This allows for users to pay for it themselves rather than having an advertiser-subsidy model like in phase 1.

Share

Conclusion

Search is being replaced by synthesis. Feeds will be replaced by curation.

For two decades, the web has trained us to either hunt for information or drown in it. The next interface breaks that bargain. It does not optimize for clicks, outrage, or time spent, it optimizes for value, relevance and balance.

If the first era of the web was about access, and the second about engagement, the third will be about enhancement. And web curators will be where that human enhancement can come from, deciding not just what we see, but what we no longer have to.

I hope to be able to contribute to this collective human endeavor before us, and perhaps build it right this time.

Join Web Curator's waitlist

The European Union is finally taking action against Big Tech.

This week’s crackdown on American platforms is long overdue, it should have happened in 2016. But while raids, fines, and public shaming make headlines, they don’t solve the core issue: We’re still stuck in a cycle of resistance without a clear vision for what comes next.

Right now, the EU is focused on what it doesn’t want: monopolistic behavior, data exploitation, and algorithmic manipulation.

But what does it actually want?

If we don’t define that, we risk replacing American tech giants with European clones, same architecture, same problems, just with an EU flag slapped on top.

Although that may sound nice, let me show you why it’s not nice enough.

The Solution: Interoperability as the New Standard

Instead of just breaking up Big Tech, Europe should enforce interoperability. Here’s what that means:

• Users own their data, followers, and identity: no platform should lock them in.

• Seamless migration: if you don’t like Twitter/X, you should be able to take your network to a competitor without starting from scratch.

• Decentralized infrastructure: social media should run on open protocols, just like the web does with HTTP. Companies should not own the identity layer, data layer and infrastructure layer. This also makes it easier to build a competitor for a thriving ecosystem.

This isn’t a radical idea. It’s how social media was supposed to work, and how Twitter worked initially

.

Subscribe now

Precedent: How the Web Won (and Social Media Lost)

In the early days of the internet, AOL controlled its own walled garden. Then Tim Berners-Lee and others built the World Wide Web, an open, interoperable system where no single company owned the infrastructure.

That’s why the web thrived.

Social media, on the other hand, became privately controlled silos. Facebook, Twitter, and TikTok don’t just host content, they own the infrastructure of interaction, your identity AND your content. That’s why they’re so hard to escape.

We need a social media equivalent of HTTP, a protocol for identity, interaction, and content that no single company controls.

Turns out it exists already….

What Would This Look Like?

1. Open Social Protocols

   • ActivityPub (used by Mastodon) is a start, but it’s fragmented.

   • Bluesky’s AT Protocol (Authenticated Transfer Protocol) is another step toward interoperability, but it’s still early.

   • Solid Project (Tim Berners-Lee’s decentralized data initiative) could be the foundation for user-owned identities.

   • Nostr Protocol is probably one of the better all-encompassing projects, but the user experience feels a bit too technical still for most people

2. Policy Enforcement

   • The EU could mandate that any platform with over X users must support interoperability.

   • Data portability laws (like GDPR) should extend to social graphs, your followers, your posts, your interactions.

   • Public funding for open-source alternatives, if the EU wants to break Big Tech’s grip, it should invest in the infrastructure to replace it.

   • Please read this policy brief from Bruegel for more detail than I could ever cover in this article. I’m just trying to make the idea more accessible!

3. Governance Without Monopolies

   • Instead of Mark Zuckerberg or Elon Musk deciding who gets banned, decentralized moderation (like community-driven rules) could work.

   • Cooperative ownership models, platforms run by users, not shareholders.

The Risk: A European Facebook

Right now, Europe is seeing new social networks pop up, but they’re just rebranded versions of the same thing. People trust them because they’re “not American,” but the architecture is identical. It’s more geopolitical rebranding or flag swapping than true progress.

If the EU wants real change, it needs to stop playing defense and start having a vision. Instead of raiding X’s offices, remove their moats. Force interoperability. Fund alternatives. Make it easier to leave than to stay.

The Future: A Coalition of Middle-Power Platforms

The goal shouldn’t be to replace Facebook with a European Facebook. It should be to create a system where no single platform dominates.

• Small, niche communities (like Mastodon instances or Bluesky’s federated model).

• User-owned data (via Solid or similar).

• Publicly funded infrastructure (so we’re not dependent on venture capital for what effectively is a public good).

Imagine if the EU created an internet in its own image? Where smaller platforms can thrive because they have good experiences. An interoperable federation of middle sized social media platforms. Setting precedent for the future where big tech platforms from both America and China would pale in comparison.

We would become the leaders in social media because our soil precisely demands to follow the most desirable architecture for netizens.

There is an inherent golden ratio between private actors and open-source projects and other digital governance models that lead to an optimal digital ecosystem. I believe getting the right organizations to build the right layers is the 21st century answer to the reductive and outdated capitalism vs communism debate which belongs in the 20th century.

Final Thought: Less Whack-a-Mole, More Construction

The EU’s crackdown is a start, but resistance alone won’t fix anything.

We need to build the alternative for refugees form big tech, one where users control their experience, where platforms compete on quality (not lock-in), and where no CEO can banish you from their aggressively expansionist digital empire.

Otherwise, we’re just rearranging deck chairs on the Titanic.

What do you think? Should Europe focus on breaking up Big Tech, or building something better? Comment with your thoughts.

(If you’re interested in curating your own ad-free, algorithm-free feed, check out WebCurator—we’re building tools for a human-centric internet.)

Also you can listen to my conversation with Kyle de Sana on the end of platform internet on my podcast:

Share

Appendix: Where to Learn More

Wait, Is this blockchain?

Decentralized social protocols don’t need blockchain, and many avoid it on purpose. Blockchains are slow, expensive, and permanently record data, which clashes with basic social needs like privacy, moderation, and the right to delete content. Social networks need fast reads, cheap writes, and flexible governance, not global consensus on every post.

By using federation, peer-to-peer networks, and cryptographic identities instead, these protocols get decentralization where it matters (identity and data control) without the cost, latency, and privacy baggage of blockchains.

1. EU Policy & Regulation

• Digital Markets Act (DMA)

  • The DMA’s official page, including its interoperability mandates and enforcement updates. Focus on Article 7 for social media interoperability and the 2026 review, which may expand rules for social networks digital-markets-act.ec.europa.eu.

• Digital Services Act (DSA)

  • Explore the DSA’s transparency and data portability rules, especially for algorithmic choice and user control digital-strategy.ec.europa.eu.

• EU Data Act (2024)

  • Details on data portability and interoperability standards for cloud and social platforms, with key deadlines in 2026–2027 hunton.com.

2. Decentralized Protocols & Open Standards

• ActivityPub (W3C)

  • The open protocol powering Mastodon and the Fediverse. Learn how it enables cross-platform interaction without a central authority brookings.edu.

• AT Protocol (Bluesky)

  • Bluesky’s decentralized protocol, designed for user-owned identities and portable social graphs. See their “credible exit” feature for how users can migrate their data arxiv.org

• Lens Protocol

  • A Web3 social graph where users own their profiles and content as NFTs. Explore their vision for user sovereignty in social media flatlineagency.com.

3. Policy Proposals & Think Tanks

• Bruegel: “Equitable Interoperability” (2024)

  • A deep dive into how the EU could enforce vertical and horizontal interoperability for social media, with concrete recommendations for policymakers bruegel.org.

• Yale Journal on Regulation: “Supertool” Interoperability (2023)

  • Argues for interoperability as a tool to break platform monopolies and empower users bruegel.org.

• EFF: Decentralization in 2024

  • Covers the rise of decentralized social media and the policy battles ahead, including public digital infrastructure as a solution eff.org.

4. Books & Essays

• “The Internet Con: How to Seize the Means of Computation” by Cory Doctorow

  • A manifesto for interoperability and user ownership in tech, with actionable ideas for regulators and builders.

• “Radical Markets” by Eric Posner & Glen Weyl

  • Explores how data as labor and common ownership of digital infrastructure could reshape platforms.

5. Tools & Alternatives

• WebCurator

  • A tool for human-centric information curation, free from advertiser influence. Built for users who want control over their feeds.

• Fediverse Apps

  • A directory of ActivityPub-based apps (Mastodon, PeerTube, Pixelfed) for decentralized social media.

• Bluesky & AT Protocol Apps

  • Explore apps built on Bluesky’s protocol, where users can port their identity and followers between services.

How to Get Involved

• Advocate: Push for stronger interoperability rules in the 2026 DMA review by contacting EU policymakers or supporting digital rights groups like EFF or EDRi.

• Build: If you’re a developer, contribute to open protocols like ActivityPub or AT Protocol.

• Switch: Try decentralized platforms (Mastodon, Bluesky) and tools like WebCurator to take back control of your digital life.

I suppose in the era of AI-slop whatever is closest to the heart seems to work its way to the centre more easily. Which is why today I want to centre the topic back to a personal matter.

Namely, as the world ruptures and breaks before our eyes, many watch transfixed. Some still under the rush of adrenaline. And some beyond the peak, fall into the uncomfortable trough of realization of what this truly means: Peaceful times are likely behind us.

Life at a greater collective level will get harder than it has been. At least for a while, before things get nicer again.

And as perhaps the reality of the moment dawns on you, I invite you to feel the feels.

This reality is immense to process, and we have absolutely no grounding in the somatic dimension in our European culture. We think we can escape what must be felt eventually.

We believe we can live in our heads, and avoid the physical reality of emotions and sensations.

I see people cumulating backlogs and backlogs of avoided emotions, like a heavyweight on their back proclaiming they are absolutely fine.

You see, the unspoken irony of suffering is that it is almost always unconscious. That is beneath the level of conscious awareness. As it arises into awareness it is already the beginning of the end of this suffering.

Although it feels like a step backwards, it’s quite the opposite.

In fact I think if people affronted their emotions fully and stoically (this is what stoicism means by the way, not some kind of dissociative performance), there wouldn’t be any wars.

And we wouldn’t go through the whole ordeal again.

And the tyrants of this world would move through their trauma before it pushed them to the impulse of some domineering grandiosity.

Jean Baudrillard aptly said in Simulacra & Simulation that fascism arises from a nation’s inability to grieve the loss of power.

By now I’m sure as you read some of these feelings may have touched you somewhat. Unless you’re a cold-hearted beast, which I very much doubt of you!

One thing which happens when strong emotions traverse individuals is people look for a channel for these emotions. Even more so when it is a collective experience!

A meaningful action perhaps.

I wanted to therefore make a suggestion on how to help in this crisis. My offer is nothing grandiose, or heroic, but rather something small.

I think there are many victims of the current rupture, however I want to point towards the most nebulous and improbable cluster of them all: The Lonely Folks.

I’m not sure if it is news to anyone, but since the advancement of the internet, social bonds have weakened in almost every way.

Sowing mistrust between people has one consequence: Isolation.

My suggestion to help is the following: If you sense that someone is lonely or socially isolated, invite them inside of your world and your crew.

One thing I have noticed since COVID is we put in place social distancing which we never really fully reverted back from. Did you not notice? There was no real official ending to the pandemic. No closing ceremony.

So culturally it simply bled into the future.

A consequence of that is that whoever didn’t find themselves inside a “folicule” or a “pod” as we used to call them, may have remained so.

Many people live far away from their families, or are estranged from them. Many more are single with numbers hitting unprecedented levels.

However there is shame attached to loneliness and I guarantee you these people do everything in their power to hide it.
So pay attention. There will be signs.

Be inclusive, hospitable in inviting people to gatherings.

There was a time where people who were lonely were in larger part perhaps people you also didn’t want to spend time with. I believe this is no longer the case.

I meet a lot of interesting, loving people that don’t seem to have anyone they can call a close friend.

Pay attention. They will make it clear. And I guarantee you that you’ve seen the signs but decided to ignore them.

My invitation is to no longer do that. And invite them to the drinks you’re organizing. Give them a chance.

We both know only regular citizens will sow back our culture together.

My godmother, who was a refugee from war used to say: The greatest danger of war is not what you think. It is not wounds. It is not imprisonment. It is the danger of forgetting how to love. Nothing is more terrifying or dangerous than forgetting how to love. Don’t allow that to happen. Ever.

Thank you for reading, as always,

AK

Share

Yesterday it finally clicked why Europe is so strategically well positioned in the new world order.

After years of constant Euro-bullying, claims that Europe and its leaders are weak, spineless, and visionless, I started to understand why this narrative kept flooding my social media: Europe is, despite appearances, the best-positioned geopolitical entity for the coming century.

This doesn’t mean the European model is perfect. It isn’t. Like any political system, it has flaws. But unlike most, it is unusually willing to admit them and course-correct. My argument isn’t that the EU has arrived, but that it is clearly moving in the right direction.

We currently live in the age of the strongman, where reality is often distorted.

Hard power, charisma, and displays of force are overrated, while quieter forms of strength are dismissed. Optics of power matter more than power itself. As a result, many people struggle to see which systems are actually working, and even align their opinions with what they think the majority believes, rather than with their own judgment.

This article is an attempt to step outside that narrative and highlight overlooked factors that suggest Europe is, in fact, in a stronger position than the United States, China, or India.

Yet the dominant story remains one of European decline, which I disagree with.

So I want to stir the conversation a little, and help rebalance our understanding of where Europe truly stands. I also believe this persistent pessimism reflects a deeper historical insecurity that Europeans need to move past, and quickly.

The EU represents a totally new kind of political entity

The European Union must be understood as a fundamentally new kind of geopolitical entity, one with no true historical precedent.

It is not “finished.” Even in 2026, the EU is still widely seen as a political project rather than a completed state. That incompleteness often gets mistaken for weakness, when in reality it reflects something else: an entity still in the process of becoming.

Geopolitical structures evolve over time through increasing complexity. Older forms don’t disappear overnight; they coexist with newer ones. But historically, the most advanced, interconnected, and economically dense regions tend to pioneer new political forms every few centuries.

We’ve seen this evolution before:

• city-states

• empires and caliphates

• feudal monarchies

• loose confederations

• nation-states

• superpower blocs (like NATO)

• supranational state (EU)

Today’s strongmen make a critical miscalculation: they believe we are returning to an age of empires. I don’t think we are.

What we are living through looks more like a transitional gap, what Gramsci described as a period where the old order is dying but the new one has not yet fully emerged. The recent reappearance of imperial rhetoric is a symptom of that gap, not the shape of the future itself. (Venkatesh Rao explores this idea well.)

Why imperial revival is unlikely to be the long-term equilibrium

1. People can organize far more easily
Mass mobilization is no longer rare or elite-driven. From Arab Spring–style uprisings to consumer boycotts and youth movements across Asia, discontent now finds coordination channels quickly. Government abuse is harder to hide and easier to resist than in previous eras.

2. Information eventually breaks through
Social media has created permanently informed populations. Even when regimes flood the system with propaganda or noise, truth tends to surface over time. It may take years, but it does surface.

3. Historical awareness is widespread
Large parts of the global population are now acutely aware of how imperialism, war, and colonization played out in the past. This makes it much harder to sell these projects as noble or inevitable.

For these reasons, I see the current age of strongmen as transitory, not as a stable new order.

Why the EU has no real precedent

At this point, a common objection arises: haven’t we seen this before? Didn’t Switzerland, Germany, or the United States go through similar phases by federating smaller units?

Short answer: no.

None of these cases involved the federation of nation-states.

• Switzerland unified city-states and rural communities.

• Germany unified kingdoms, duchies, and principalities.

• The United States unified colonies that were not independent nations in the modern sense.

This is a fundamentally different governance challenge.

Connecting rural communities or principalities is not the same as integrating nation-states the size of Germany, France, or Spain, each with its own language, army, diplomacy, and historical identity.

The EU is the first mechanism in history where nation-states voluntarily pool sovereignty at scale, not as a temporary alliance, but as a durable system. What began as a trading block and peace project is gradually evolving into a genuine geopolitical actor, capable of rivaling traditional nation-states like the United States and China.

That is why the EU looks strange, slow, and unfinished when judged by old standards, and why those same standards consistently fail to measure its actual strength.

Subscribe now

Proof that Europe is strengthening

Europe technically has the world’s largest army

I won’t even start with soft power here, despite it being wildly underrated and an area where Europe clearly excels.

If the United States continues its trajectory of disengaging from NATO, the most likely outcome is a NATO structure centered on Europe rather than Washington.

EU member states alone account for roughly 1.6 million active military personnel. If Ukraine joins in the coming years, that number rises to around 2.5 million. Including the remaining NATO members brings the total close to 3 million.

That compares to:

• China: ~2 million

• United States: ~1.3 million

• Russia: ~1 million

You might argue there is no such thing as a “European army”—and you’d be right. For now. But integration, coordination, and joint procurement are clearly moving in that direction. In the long run, Europe is structurally strong in military manpower, even before accounting for its industrial and technological base.

Europe has built the world’s largest free trade zone

The EU has recently finalized or advanced major trade agreements, most notably with Mercosur and India, while deepening common procurement and industrial coordination with partners like Canada.

More broadly, Europe has become the central node of the world’s most extensive alliance and trade network. No other major geopolitical actor is as consistently capable of striking large, balanced agreements with other major economies.

These are not extractive deals. They are designed to be durable, mutually beneficial, and rules-based, exactly the kind that compound influence over decades.

It’s the only bloc countries actively want to join

Name another major geopolitical entity that states actively seek membership in.

Taiwan isn’t eager to “join” China.
Greenland isn’t lining up to become American.
Ukraine certainly didn’t want Russia.

Yet countries continue to reform, wait, and endure political pain to join the European Union.

That alone says something profound about the EU’s gravitational pull.

Europe leads on the societal indicators that actually matter

Education quality, healthcare outcomes, social mobility, and income inequality, across these metrics, Europe consistently ranks at or near the top.

These are leading indicators of long-term resilience, productivity, and social stability. Power that cannot reproduce itself internally eventually collapses. Europe is unusually good at maintaining its foundations.

Europe is the world’s most reliable bloc

Reliability is an underrated form of power.

Europe honors contracts, maintains regulatory continuity, and changes policy slowly and predictably. This is precisely why:

• The euro (and Swiss franc) are the most credible reserve-currency candidates after the dollar

• Capital fleeing instability in this era often flows into Europe, not away from it

• Middle powers increasingly see Europe as the safest long-term partner

In a volatile world, consistency becomes strategic advantage.

China’s power has limits, and the U.S. is bleeding credibility

China is a serious power. Its GDP rivals that of the EU, it has vast industrial capacity, and its global supply-chain reach is real.

But China struggles badly with soft power. Its diplomatic toolkit relies heavily on pressure and leverage, which works, until it doesn’t. Over time, this approach alienates middle powers rather than binding them.

China often appears stronger today largely because U.S. soft power has deteriorated so sharply by comparison. That contrast won’t hold forever.

Any military move on Taiwan, in particular, would severely damage China’s global credibility and accelerate strategic pushback, regardless of the outcome on the ground.

The pattern is clear

Europe doesn’t project power loudly.
It doesn’t dominate headlines.
It doesn’t fit 19th-century definitions of strength.

But across military capacity, trade, alliances, societal resilience, and trust, the trend is unmistakable: Europe is consolidating power while others burn through it.

Why the supranational model is the future of political organization

The supranational model exists because middle power sovereignty is more at risk than ever.

In a world of shared supply chains, capital flows, climate risk, and security dependencies, no nation is an island. The real question is not whether sovereignty is limited, but whether it is lost to larger powers or pooled voluntarily into a collective.

This dichotomy is clearest in the case of Brexit. As a middle power (that thinks it’s a large one), it can either be under the US or the EU umbrella, and it clearly choosing the latter

The supranational model is built on collaboration rather than domination. By design, it connects rather than absorbs, and in geopolitics, being the connector is often the strongest position to hold. Instead of a single hegemon, it allows for plural sovereignty, where no state is conquered, exit remains possible, and power is negotiated.

It is also a trust-based architecture. Former enemies are reconciled through permanent interdependence, not force. This is rare in history, and unmatched at this scale. The system is pragmatic and iterative: it adjusts, regulates, experiments, and corrects course as conditions change.

Populism thrives where there is no clear long-term direction.

Europe’s mistake has not been the absence of a vision, but its reluctance to articulate it and for Europeans to acknowledge it. The supranational model is that vision: a quiet, institutional answer to governing complexity without empire.

The moment has come to own it, support it, improve it, and actively contribute to it, rather than judging it by outdated models of power.

Share

We should imagine a new kind of digital life for ourselves while the window to change is open.

One that is healthier. One that works for us. One that leverages the power of AI to make us better citizens, with agency. Not a continuation of the destructive era of platform internet.

As part of my new information diet I have pledged to remove all traditional feeds. This means, no infinite scrolls, no information curated for me by a black box, no ad models.

In this article I want to discuss why the newsfeed as the central space of the internet has been so detrimental and what we can create instead.

A brief history of the newsfeed

Facebook launched the Newsfeed in 2011. I’m not sure if anyone remembers but until then the UX architecture was profile-centric.

Then progressively, following its implementation, the newsfeed became more central and more algorithmically powered until by 2015, a machine-learning algorithm was deciding most of what people were seeing, and became the main digital space.

Then, as if per coincidence, in 2016 Brexit and Trump happened.

You can virtually trace these technologies back to the enabling of populist movements. One cannot exist without the other. Populist politics are an inevitable symptom of a society based on feeds.

I cannot stress enough that as long as we have newsfeeds, we will have extremist populist governments that are great at criticizing institutions, pointing fingers and dismantling world orders, but cannot create, ally, build or get anything of quality done.

Which is why the sooner we build better alternatives, the sooner we are out of this weird chapter of history all of us are hoping will end soon.

This article describes the history of Facebook in more detail if you’re interested. A timeline below helps make sense of it visually.

Subscribe now

Why are feeds so bad?

1- It deregulates your dopamine pump and messes with your attention

The information diet is very similar to your food diet in many ways. One of those is that if you keep eating low nutritional value, you end up deregulating your system.

Now the science is still not fully conclusive on the fact that social media or dopamine deregulation can actually cause ADHD directly. But I have a suspicion it is very likely. I have myself developed symptoms over time. and am fairly certain I can attribute it to my scrolling habits.

Pseudo-scientific speculations aside, do you really want to wait for proof to come out once you may have already acquired irreversible ADHD symptoms or would you rather play it safe right now and get off them?

That’s what I thought.

2- It amplifies the worst parts of you

Newsfeeds are designed to make you go into an unconscious semi-slumber.

Once you are in that state it is much easier to prick you on your rage or fear nerves. In the same way that when you are tired it’s easier to get you angry or afraid. Because you are less “there” to regulate yourself.

What happens is it trains you to react in the most unconscious ways, watering your seeds of fear, of anger, of suspicion and any other emotion where you outsource agency.

We all know this at some level, but naming it helps.

3- It brainwashes and polarizes you

It has been proven that social media has the capability of radicalizing an otherwise moderate individual. The ability for the tools to corner you into a narrative and extrapolate your small fears into a group of people that will say that these outcomes are “very likely” will inevitably distort your view of the world.

In a way there is no such thing as shared reality however when collectively there is little overlap, it becomes harder to function as a society.

So, what can we replace newsfeeds with?

I think I have made it very clear that feeds are optimized for value extraction for advertisers and for the platforms themselves much at the detriment of the users. Hopefully after reading this, you also sense that you deserve better.

Now what should come in its place?

Certainly something that users pay for, there is no way around it. If you are subsidized, you are utilized. This motto in tech should be very clear by now.

I have been thinking at great length as to what different roles feeds have in my life, and what are better replacements:

1. Get a break from stuff / dissociate → Go for a walk, listen to some music, read a book

2. Get some light-hearted entertainment → Watch a movie, socialize, play with my dog

3. Keep up with my friends lives → Call them? Hang out with them?

4. Stay informed about the world → Get a quality subscription somewhere, where curation serves me

5. Know what’s going on in my industry → ?

6. Learn about new hobbies → ? (Youtube tutorials are pretty unbeatable tbh)

7. Meet like-minded folks → Go out and socialize!

8. General undefinable FOMO → Go out and socialize!

When it comes to the social stuff I’m pretty sure that we will find ways to be social without social media given how lousy a job it has done to solve any of this.

All the entertainment-related stuff, you just have to find healthier alternatives.

When it comes to keeping up with hobbies, industry news and general info I definitely saw a gap there.

That’s why i started building: webcurator.io

I can see how a lot of people seek to curate information for their info diet, and get lost in the swamps of feed distraction. Given the fact that the unit economics now allow for individuals to have their own web crawlers and curators, I decided it was time for netizens to start paying for their own curator services, and not outsource that to an advertiser subsidized feed.

I am in the early days of building this tool, but am very keen on feedback from you guys. I have received lots of encouragement already.

Currently Web Curator is still in Beta and am slowly accepting people from the waiting list that you can sign up for at this link. I would rather have users that care about fostering a good community around this, and making this solution as good as possible.

I am also convinced that if my solution is the first of many that will emerge.

Let’s allow for AI to curate information for us rather than curate us for AI, which is the world we exist in today.

It’s time to make internet human-centric again.

…

Warmly,

Karim

Share

In 1930, my great-grandfather and his immediate family fled Budapest for London. He left behind his family name, his religion, his friends, his entire life. It was a brutal upheaval, and my family still bears the scars. Yet those who stayed often fared far worse.

Today, we face another kind of global geopolitical rupture, one that will again force mass migration. But for the first time in human history, the exodus will be digital as well as physical.

Just as continental Europeans once sought refuge from the Third Reich, Europeans may soon need to escape the reach of a MAGA-stack, which increasingly views Europe as a threat to its model.

The coming digital migration, likely the largest in history, will not be voluntary for the most part. The EU’s Anti-Coercion Instrument (ACI), designed to cut off access to Europe’s market where it hurts most, will inevitably target American big tech eventually, the (only) engine of US growth. Trump’s escalation is all but certain, and Europe might have to deploy this weapon at the worst possible moment.

If this happens, which is not a given, the risks are severe. Companies could lose their core business drivers overnight. Individuals might be cut off from friends, networks, and essential data. I may sound alarmist; but this is a plausible scenario, and one we must prepare for.

In this article, I will outline how to navigate this digital migration, how to safeguard what matters and avoid worst case. It builds on my previous piece, but the stakes already feel higher.

Also you can read this interview I did for the European Correspondent on the matter.

Principles for a smooth digital migration

1. Establish Your Principles: Value-Based Realism

I borrow this framework from Alexander Stubb (President of Finland): value-based realism.

It means anchoring your decisions in your core values while confronting reality as it is, not as you wish it to be. It’s a kind of informed vigilance. Which is neither naive, nor paranoid. Apply this lens to every part of your digital stack. When selecting tools, ask yourself:

Key Risks to Evaluate:

1. Privacy: Who can access my data?

2. Brainrot: Does this tool degrade my focus or mental stability?

3. Manipulation: Am I vulnerable to algorithmic or political influence?

4. Key information Loss: Could I lose everything tomorrow? How would that affect me?

Critical Questions for Every Tool: What to ask when looking

• Where is the company based? (Jurisdiction matters.)

• What is their business model? (Are they selling your data or attention?)

• What is their governance model? (Who controls the platform? Open source? Non-profit, for profit? What are the chances of this changing?)

• How do they use my data? (And what could a rogue regime do with it?)

• Would I feel comfortable sharing this data with a stranger? (If not, why trust a corporation?)

• What if this data disappeared tomorrow? (Could you recover?)

• How do I feel after using this app? (Drained? Empowered? Manipulated?)

• How encrypted is my data? (End-to-end? Zero-knowledge? Or just marketing fluff?)

2. Embrace friction: Increase digital agency

Friction is your path to agency.

Once you push through inconvenience once, you can do it again with less resistance. The mind strangely lights up when challenged in this way. Yes, it feels scary, but make the jump.

How to Start:

• Ask questions. Use LLMs to fill gaps in your knowledge on current and alternative tools. Stay informed.

• Boost your digital literacy. Understand concepts like end-to-end encryption (E2EE), business models, and layers of the tech stack.

• Be an informed citizen, not a passive user. Knowledge is power, and power is resilience.

Remember: It’s harder to capture an agile animal than an idle one.

Subscribe now

3. Decoupling: A Multi-Speed Approach

Not all tech deserves the same exit strategy. Below, I’ll help you determine the right way to decouple from a tool or ecosystem based on its value, or lack thereof.

Category 1: No Value Added: Delete Now

If a platform is a net negative, draining your time, attention, or energy without tangible benefits, you know you can delete it immediately. Yes, it feels strange at first, but the relief afterward is undeniable.

Emotions will surface; give yourself time to process them. Don’t underestimate the somatic toll of decoupling, but trust that it’s worth it.

Category 2: Strong Entanglement: Decouple Slowly

For deeply embedded tools (e.g., Google stack, WhatsApp), transition gradually:

1. Transfer essentials to your new platform (e.g., Proton Mail, Signal).

2. Monitor the old account for 6 months to catch anything missed.

3. Delete it once you’re confident nothing critical remains.

Category 3: Value with Boundaries: Restart on Your Terms

Keep tools that add value, but reset your relationship with them. For example:

• Delete your old account and restart fresh with stricter privacy settings.

• Use ChatGPT for utility, but agree with yourself: no personal data, no emotional dependency.

4. Replace Intentionally

This is the step that excites me most as an entrepreneur. It’s where you start to see the gaps in the market, not just for yourself, but for others who are also seeking better alternatives.

When I left traditional social media feeds, I realized I still needed a way to curate information. That’s why I built webcurator.io: a tool designed to fill the void left by platforms I no longer trusted or wanted to use.

How to Approach Replacement:

1. Visualize Your Needs Ask yourself:

   • Why do I actually use this platform? (Is it habit, necessity, compulsion or something else?)

   • What would a healthier alternative look like? (What features would it need? What would it exclude?)

   • Does an alternative already exist? (If not, is this an opportunity?)

2. Consider Building It If the right tool doesn’t exist, ask: Should I build it? The best solutions often come from personal pain points. If you’re feeling the gap, others likely are too.

Share

Keep it coming folks your encouragement on this topic has meant the world! Reach out if you want to discuss further.

Karim

After publishing my last article on how migrating from big tech platforms as a user is easy as pie (link below), I realize we were in the middle of a seismic tech shift.

People are fed up with Silicon Valley as our global kings, and the people are asking for better systems! Sarcastic Marxist sentiment aside, I seriously believe the floodgates to the internet are reopening, and there is an opportunity for a post-American internet.

For more on why now is the moment, watch this talk below, by activist Cory Doctorow:

But for now I want to explain to you how platforms have gatekept the internet, so we can counter that now faltering strategy.

The Problem: Feeds Are Quicksand for Your Mind

While redesigning my information diet, I realized something brutal: As long as I let platforms control what I see, I’m screwed.

Social media feeds aren’t just biased, they’re designed to erode your agency. Imagine landing in a new city where your tour guide owns all the restaurants and **hypnotizes you into eating what they want, not what you want.

That’s what platforms do. They don’t just curate, they manipulate. They turn you into a reactive, dopamine-addicted zombie who scrolls instead of chooses.

So I decided: I’ll build my own feed. A curation layer on top of platforms, so I can discover great content without outsourcing my attention.

But then, unsurprisingly, I hit a wall. Four terrible options, each with its own catch.

Subscribe now

The Four (Bad) Ways to Access Platforms

(And Why They All Suck)

1. RSS Feeds: The Ghost Town

What it is: RSS lets you subscribe to updates from specific creators or sites. Some platforms (YouTube, Reddit, podcasts) still offer it. Others (Meta, TikTok) killed it years ago.

The Problem:

• No discoverability. RSS only works if you already know who to follow. What if the best content comes from someone you’ve never heard of?

• Fragmented. Every creator has their own feed. Good luck stitching together a coherent experience.

• Outdated. Platforms want you stuck in their apps, so they deprioritize RSS (or break it entirely).

Verdict: ✅ Works for niche use (e.g., following a handful of creators) | ❌ Useless for discovery.

2. APIs: The Paywall

What it is: Platforms offer APIs (Application Programming Interfaces) to let developers pull data, if you pay up.

The Problem:

• Insanely expensive. APIs are priced to lock out indie devs. YouTube’s API, for example, costs $0.002–$0.02 per request. Scale that to thousands of users? Bankruptcy city.

• Rate-limited. Platforms cap your access to prevent competition. Hit the limit? Locked out.

• Unreliable. APIs change or break without warning. Build on them, and your tool dies overnight.

Verdict: ✅ Works for personal scripts | ❌ Impossible to scale.

3. Anonymous Browsing: The Cat-and-Mouse Game

What it is: Use tools (VPNs, burner accounts, scraping bots) to sneak into platforms without detection.

The Problem:

• You will get caught. Platforms have anti-bot systems (CAPTCHAs, behavior analysis). Scale up, and they’ll ban your IP or sue you.

• Ethically sketchy. Even if you’re “just” building a better feed, platforms see it as theft (lol).

• Unsustainable. Works for personal use, but try sharing it with others? Lawyers at your door.

Verdict: ✅ Works for personal, small-scale use | ❌ Legal and technical minefield at scale.

4. Crawling: The Whitelist Wall

What it is: Write a bot to scrape platform data directly, bypassing APIs.

The Problem:

• Whitelists. Platforms block scrapers by default. Only approved (read: corporate) crawlers get through.

• Legal threats. Scraping violates most platforms’ Terms of Service. They will sue if you gain traction.

• Technical arms race. Platforms constantly update their anti-scraping tech. Your bot breaks weekly.

Verdict: ✅ Works for research (if you’re sneaky) | ❌ Dead on arrival for public tools.

Cool hack: If you want to check which crawlers are allowed on a site, check https://domainurl/robots.txt. E.g. to check who is allowed to crawl Facebook go to https://facebook.com/robots.txt - Interesting right? This way you know who’s doing business with whom.

The only possible workaround - smash the gatekeeper’s pact

The pact: Gatekeepers only whitelist each other

You get it now. There’s no way to access platform content on your terms. Want to see my content? Fine, but only through my enshittified feed. The CEOs of Big Tech didn’t just compete, they colluded. They sat in a room and said: “If you whitelist my content, I’ll whitelist yours. Together, we’ll build gated citadels, and hoard all the world’s knowledge and attention inside.”

Everyone else? Locked out.

This isn’t just in tech, it’s how every oligarchy works. But here’s the twist: In tech, the tide is turning. People are sick of the incumbents, and global decoupling means Europe has a shot at building something healthier. More on why in my article below:

The Solution: A Fragmented, Interoperable Internet

It’s simple (if radical): We need a web where no five guys in Silicon Valley control the gates.

Imagine if:

• More creators hosted their own content (on smaller, open platforms).

• More curators could freely crawl and remix that content (like the early web!).

• Power was distributed, so no single platform could dictate what you see.

That’s the internet we lost. And the one we can rebuild but differently.

When Google launched, nobody blacklisted its crawlers. Sites were happy to be discovered, because it meant more traffic, not less. What I’m proposing? A return to that spirit, but with more diversified and accessible tools.

How? Three Steps to a European Web

1. Post the good stuff on Open Platforms

• Use ActivityPub (Mastodon), Bluesky, or self-hosted blogs.

• If you build your own, Blacklist OpenAI/Google crawlers if you want, but welcome crawlers from the open web.

2. Build Open Curators (this is what I wanna do)

• Create web-crawling tools that highlight de-platformed content.

• Example: A “European Discovery Engine” that indexes only open, interoperable sites.

• A trusted gateway to the web you pay for that curates content for you

• If this is something you want to build, contact me.

3. Starve the Oligarchy

• Stop feeding the beasts. Every hour you spend on Facebook/YouTube/TikTok funds their gatekeeping.

• Pay for your own tools. Unlike the early web, today’s tech is cheap enough that citizens can own their infrastructure, no corporate subsidies needed.

The Stakes: Oligarchy vs. Democracy

The American internet is an oligarchy. The European internet won’t be.

The old system emerged when tech was expensive, so only corporations could build it. Not anymore. Today, the unit economics work for us. We can pay for our own web.

So let’s build it. An internet that works for citizens, not Silicon Valley.

What’s your plan to escape the gated citadel? Building a tool? Switching platforms? I would love to hear your ideas in the comments.

Share

The most enshittified = cognitively harmful, agency-robbing and privacy violating platforms.

Cutting out the worst part of your diet is always the first step towards health. This applies to food (sugar, processed food, fatty oils, etc) and digital diets are no different. This is my first step towards a healthy information diet (read my article below).

This article is meant to be read as an off-ramp to the digital prison we have all chosen to stay inside of, showing that there is another way, and it’s really not that hard to do!

Why Did I Do It?

I spent 20 years on social media. I joined Facebook in 2006, fresh off a summer at Phillips Exeter Academy outside Boston. Back then, it was the perfect way to stay connected with friends scattered across the globe, a separate, global social layer beyond just the people physically around me. Oh, the innocence of those early days!

But over time, these platforms morphed subtly, like a pot of water slowly coming to a boil. And today, I realized: the temperature is just too high.

Three Reasons I Left:

1. My mental health was visibly declining.

   A growing body of research shows that not all platforms are created equal. Some, like WhatsApp and Snapchat, have neutral or even positive effects on well-being, self-esteem, and friendship closeness. Others: Instagram, YouTube, and especially TikTok, are linked to negative mental health outcomes. I felt it myself: more anxiety, less focus, and a constant sense of unease and my dopamine-driven brain and compulsive behavior kept increasing.
   

   The caveat, it took a lot of energy to actually NOTICE how that was happening. Curious how this stuff can be kept from our own awareness for so long.

   

2. I felt manipulated, confused about what was real, and lost my sense of agency. Turns out, the “tinfoil hat” crowd wasn’t entirely wrong. The algorithms, the endless scroll, the curated realities, it all adds up to a loss of control over my own attention and perception.

   Mechanisms of these platforms are such that they create an enclosing “digital sphere” where all the reality and narrative creation comes from one entity. Breaking out of your echo chamber is more easily possible in a fragmented environment with different players giving you different views. Nobody should hold narrative monopoly.

   

3. Reliance on U.S. Big Tech started to feel unsafe.

   For years, I’ve been uneasy about the world’s most powerful organizations knowing more about me than my own mother or best friend.

   If you’ve seen how surveillance states operate, think East Germany’s Stasi, you’ll realize we’ve willingly invited those “mics” into our homes. The difference? We chose to do it.

   But the threat on Greenland was the redline for me.

   The American State threatening European sovereignty and to invade European soil means they will stop at nothing, including great power clash, and radically cannot be trusted.

   There is really no reason for the American State to be able to design an extremely intimate profile of who I am when that trust is by no means deserved. After strong reflection it felt like a violent boundary crossing.

   I imagined an American spy getting a download on my favorite music songs and recipes and being able to convince me to go on a date with them, and it made me shiver.

Subscribe now

How Did I Do It?

1. I Deleted All “Enshittified” Social Media

Any social media with a feed, where I end up mindlessly scrolling, had to go. Especially platforms focused on short-form content or AI-generated sludge. These are the dopamine-crazy ones, designed to hijack attention and erode agency.

I requested to download my data, then permanently deleted my accounts on:

• X (Twitter)

• Facebook

• Reddit

• Instagram

• Character.AI

• TikTok

• Hinge (and other dating profiles)

Thank God for GDPR and the right to be forgotten. I don’t know how long the EU can enforce these rules since international law is eroding by the minute, but I’d rather remove my traces while I still can. The sooner I’m off these platforms, the longer the gap where they don’t know what I’m up to.

What I didn’t delete (but committed to using less):

• WhatsApp: I’m migrating my intimate and priority conversations to Signal. WhatsApp is harder for me to access (notifications off for now), which helps reduce compulsive checking.

• OpenAI, Claude, and other American LLMs: I’m exploring private, ethical alternatives. This is arguably the most intimate tool of all. Really no reason for Sam Altman to know which insecurities I have at 3am.

• LinkedIn: I still need it for networking and finding clients, but I’m using Typefully to schedule posts and limit my exposure.

• Google Suite: More on this below.

2. I Replaced Google Suite with Proton

What I store on Google Suite is some of my most precious work information: Gmail, Drive, my work calendar, Google Meet, where all my meetings happen (does Google record a transcript?).

I don’t want this critical data leveraged by an organization that still fundamentally operates on an ad model. Google knows everything about my work and could nudge me in ways I don’t even realize.

Proton is great! Trustworthy, based in my native Switzerland, with servers physically located beneath their HQ. No shady data practices, no ads, no hidden agendas.

Next step: I’ll be using Proton’s login masking tools so I don’t get sold ads when logging into random apps and websites.

Also I want to get to the point where I can delete my Google Account. But for now I am just transferring dependency and keeping it as an archive.

Highly recommend Proton! (Use my referral link)

3. I Downloaded Opal to Block Distractions

I installed Opal and immediately blocked all news sites.

I’d rather have scheduled, intentional touch points to keep up with current events, from platforms that are analytical, thorough, and not in the business of emotional manipulation (“Here’s what happened… and here’s how you should feel about it!”).

I also use Opal to limit usage of platforms I still need but want to keep at arm’s length. If you struggle with digital distractions, try it, it’s fantastic.

4. I Invested in High-Quality, Long-Form Content

If I’m carving out digital space in my life, I need to replace it with something better. Mostly, I’m focusing on IRL experiences, but I also need a high-quality information diet.

So I:

• Subscribed to The New Yorker.

• Got paid subscriptions to my favorite Substack writers.

• Bought a ton of books I’ve wanted to read for years but couldn’t because my attention span was shot.

The difference is extremely noticeable. After cutting out social media and replacing it with long-form content, I feel more regulated, more present. You can feel it in your body, the absence of that jittery, scattered energy.

How Does It Feel the Morning After?

Joy. Pure, unfiltered joy.

I don’t remember feeling this light since before the social media era. It’s like a weight has been lifted, one I didn’t even realize I was carrying.

The mental fog has been clearing from this and all the effort I have put in the past month, and for the first time in years, my mind feels like mine again.

But this is just the beginning. There’s more to do, and I’m resolved to see it through.

What’s Next?

1. Login Masking

I’m going to use Proton’s login masking feature so I can log in with a different masked email every time. This way, my user profile for ads isn’t tied to a single email, no more centralized profiling, no more creepy personalization.

2. OpenAI, Claude, and Other American LLMs

I’m committing to deleting my profiles on these platforms this year, once I’m comfortable with a private, ethical alternative. I refuse to feed my data into systems that prioritize profit over privacy.

3. Phone: The Fairphone Experiment

I’m seriously considering buying a Fair Phone this year. It’s 1/3 the price of an iPhone and probably 80% as good, a fantastic trade-off compared to being spied on and locked into Apple’s repair monopoly.

4. Laptop: Seeking Alternatives to Mac

I don’t know of good alternatives to Mac yet, but I’m open to suggestions! Long-term, this is going to be critical. If you know of ethical, privacy-focused options, let me know.

5. Browser: Ditching Chrome for Good

I’m on Arc right now, which is still Chrome-based. Eventually, I want a browser that works for me, not for them. Firefox? Brave? I’m exploring.

6. Google Maps: The Last Frontier

This one holds critical, sensitive data, where I go, who I meet, my routines. I’m still researching alternatives. OpenStreetMap? Something else? I need a real replacement.

7. Amazon / Temu: Low Priority, But Still a Problem

They’re not harming me as directly as social media, but I’d still rather live in a world where neither of these platforms existed. Decentralized marketplaces, anyone?

What I’m Choosing to Keep (For Now)

I’m **limiting—but not eliminating—**my use of:

• YouTube (for long-form content, though the algorithm is getting worse).

• Substack (for high-quality writing, despite the “Notes” feature creeping in).

• Spotify (for podcasts).

Why? Because they still offer long-form, high-quality content, something I’m not ready to give up entirely. YouTube is following TikTok’s enshittified path, and Substack’s “Notes” feature is a slippery slope, but for now, I need these tools to stay informed for my work.

This points to a gap in the market. I want a platform where I can access high-quality, curated content without stepping into the quicksand of attention-numbing newsfeeds.

This is why I started building this myself (shameless plugin I know), it’s called Web Curator. Your own lense into the web. Gathering folks for the waiting list here. Join in on the party!

The Lingering Fear of Missing Out (FOMO)

Yes, there’s still FOMO. What if WW3 breaks out and I miss it? What if I disconnect from friends in Asia and lose touch forever?

This morning, I chose to sit with those fears. I even did a little ceremony of deletion, pausing to feel everything that came up as I hit “delete” on each account. It wasn’t pleasant, but it was insightful. I recommend!

I know there’ll be a transition period, a bit of discomfort before the new habits settle in. But I’m holding onto that empty seat at the table, trusting that the right people and the right information will find their way back to me.

The Bigger Picture

We don’t have to remain enslaved to these platforms. We deserve a better internet, one that respects our attention, our privacy, and our agency.

I think this is the start of Internet 3.0, and I’m happy to already be a part of it.

You have your exit ramp now…

SO, what are you waiting for? Get off the brain rot train!

Share

Alternative Tools for a healthier information diet

1. Privacy-Focused Communication & Social Media

• Signal – Encrypted messaging (replaces WhatsApp, Facebook Messenger).

• Mastodon – Decentralized, open-source social network (replaces Twitter/X).

• Lemmy – Reddit alternative, community-driven and open-source.

• Web Curator - the new lens into the web, curating a feed you build yourself, intentionally

2. Email & Productivity (Replacing Google Suite)

• Proton Mail – End-to-end encrypted email.

• Proton Calendar – Secure, privacy-focused calendar.

• Proton Drive – Encrypted cloud storage.

3. Browsers (Replacing Chrome, Arc, Safari)

• Firefox – Open-source, privacy-focused browser.

• Brave – Blocks ads and trackers by default.

• LibreWolf – Hardened, privacy-focused Firefox fork.

4. Maps (Replacing Google Maps)

• OpenStreetMap – Community-driven, open-source maps.

• Organic Maps – Offline maps with no tracking.

• Magic Earth – Privacy-focused navigation.

5. Distraction Blocking & Focus

• Opal – Blocks distracting websites and apps.

6. Ethical Hardware

• Fairphone – Modular, repairable, and ethically sourced smartphone.

• Purism Librem – Privacy-focused laptops with hardware kill switches.

• Framework Laptop – Modular, repairable, and upgradeable laptop.

7. AI & LLMs (Replacing OpenAI, Claude)

• Mistral AI – Open-source, ethical AI models.

• Hugging Face – Open-source AI community and models.

• LocalAI – Run open-source AI models locally on your device.

• Lumo - Proton’s privacy focussed LLM, pretty decent!

Resources for Further Exploration

• European Alternatives to Big Tech: european-alternatives.eu

• Typefully: typefully.com (for intentional social media posting)

• PrivacyGuides: privacyguides.org (comprehensive privacy tool recommendations)

• Gloria Mark’s work on digital mental health: read these papers so you have no excuse and cannot say you don’t know what these feeds do to you. scholar.google.com

Setting the stage: beyond “social media is bad”

People often say social media and the attention economy at large is bad for you. It leads to cognitive decline. It weakens your attention.

Yet I rarely here people explain express the mechanics behind it. This is a shame because unless we understand the how, we can never fix it.

In this piece, I want to focus on one specific, and largely unexamined, mechanism of the attention economy: task derailment.

I want to explain how it works, why it is so effective, and how simply recognizing the pattern already weakens its grip. From there, we can talk about how to rebuild something many of us feel we’re losing: cognitive agency

What do I call task derailment?

It is the hijacking of your attention in a way that redirects you toward a different goal than the one you originally chose before completion.

I want you to imagine a street car, going towards an initial destination (the yellow star). Now a foreign entity steps in and puts in a rail fork that derails the street car to go towards the blue star.

Now imagine how much time, fuel, and energy it is going to take the driver of this street car to go towards the yellow star if the derailing keeps happening.

This is exactly what is happening to your cognitive function when you are trying to perform deep work.

Task derailment undermines deep work

Deep work is a foundational pillar of the knowledge economy.

Any activity that produces durable intellectual value, whether that is writing, research, programming, strategy, or design, depends on sustained attention applied over time.

Deep work is the mechanism through which cognitive labour compounds.

Task derailment interferes with this process in a subtle but systematic way. Deep work requires continuity. This means, you set a goal, you reach it. You set a goal you reach it. And repeat!

Attention must remain aligned with a single goal long enough for complexity to unfold and for insight to emerge. Task derailment does not eliminate attention, it redirects it. The work continues, but the trajectory shifts before depth has had time to form.

What makes this particularly corrosive is that derailment does not depend solely on external interruption. Research on attention consistently shows that people continue to interrupt themselves even when notifications and external triggers are removed. After prolonged exposure to fragmented digital environments, self interruption becomes habitual and eventually embodied.

This helps explain a finding that initially appears counterintuitive. When social media or messaging tools are removed, people who are otherwise highly self regulated sometimes struggle more rather than less. Many had learned to rely on frequent micro breaks to manage cognitive load. When those breaks disappear, they are left with an attention system that has not been trained to rest or recover without stimulation.

Tools that reduce interruption can help at the margins. Disabling notifications, batching communication, or using focused work modes lowers the frequency of derailment and makes sustained effort more accessible. These interventions are useful, but they remain entry points rather than continuous solutions.

Large technology platforms are not neutral spaces. They are optimised systems that reward redirection rather than completion. Their economic logic favours frequent shifts of attention over sustained alignment with a goal. Over time this produces a gradual erosion of agency, not through coercion, but through repeated, rewarded diversion.

The asymmetry is difficult to ignore. Individuals must exert continuous effort to preserve their capacity for focus, while platforms benefit from its steady weakening. Task derailment emerges from this imbalance, not as a personal failing, but as a predictable outcome of the environment.

Deep work is often the first thing to suffer as a result.

Three levers for reclaiming agency

1. Increase intentionality and structure around deep work

Put barriers around your rails so no one can highjack them

Task derailment thrives when the destination of attention is weakly defined and the conditions of work are left implicit. Increasing intentionality means making the trajectory of the task and the regulation of effort explicit before work begins.

This can be supported through a small number of structural choices:

• Define a clear and specific goal for the session and write it down in a way that remains visible

• Clarify what completion looks like for that session, even if it represents partial progress

• Set a fixed time boundary using a timer so duration does not need to be negotiated internally

• Shape the environment to reduce latent pull, for example by placing the phone in another room

• Schedule breaks in advance and treat them as part of the work structure rather than as interruptions

When effort, rest, and duration are explicit, attention no longer needs to seek relief on its own. Much of what appears as distraction is the nervous system attempting to regulate unmanaged load. If you want to learn more about this, Cal Newport is the best expert I found on the topic.

2. Train attention rather than only protecting it

Attention is a muscle and it should be trained

Structural changes reduce exposure to derailment, but they do not strengthen the underlying capacity or muscle to remain with a task. Training attention changes the relationship to boredom, discomfort, and uncertainty, which are often the moments where derailment begins.

This training can take several forms:

• Practicing sustained single task work when stimulation drops

• Sitting with boredom without immediately resolving it through novelty

• Engaging in contemplative practices that develop sustained attention, such as breath focused meditation or Vipassana

Over time, this reduces the perceived threat and discomfort tolerance of low stimulation and increases confidence in one’s ability to stay aligned with a chosen goal.

But don’t overdo it. Attention capacity compounds slowly.

3. Develop meta awareness of interruption patterns

Observe the mechanics of highjacking

Meta awareness is the capacity that makes the other levers effective. Preventing interruption matters less than noticing it as it happens and understanding the conditions that produced it.

This involves:

• Noticing the moment attention diverges without judgement

• Guiding attention back to the task deliberately rather than forcefully

• Observing how diversion feels in the body and mind

With repeated observation, patterns emerge around fatigue, uncertainty, and emotional resistance. Recognising these patterns allows regulation to be built in deliberately rather than outsourced to distraction. The more you understand your patterns, the more intentionally you can reshape them.

Share

Subscribe now

Sources and deeper recommended reading

Attentional fragmentation & interruption costs

• Mark, Gudith & Klocke (2008) — The Cost of Interrupted Work: More Speed and Stress (CHI).
  Open PDF: ics.uci.edu | ACM/DOI page: ACM Digital Library

• Mark et al. (2016) — Email Duration, Batching and Self-interruption… (email patterns ↔ productivity/stress).
  Open PDF (Microsoft Research): Microsoft

• Mark, Iqbal, Czerwinski et al. (2017) — How Blocking Distractions Affects Workplace Focus and Productivity (Ubicomp adjunct PDF).
  Open PDF: interruptions.net

Goal shielding / goal competition

• Shah, Friedman & Kruglanski (2002) — Forgetting All Else: On the Antecedents and Consequences of Goal Shielding (JPSP).
  PDF (Ovid): Ovid | Alt landing page: ResearchGate

“Unfinished goals” / Zeigarnik lineage

• Masicampo & Baumeister (2011) — Consider It Done! Plan Making Can Eliminate the Cognitive Effects of Unfulfilled Goals (JPSP).
  PubMed: PubMed | Open PDF: users.wfu.edu

Variable reward / dopamine: why feeds outcompete long-horizon goals

• Schultz, Dayan & Montague (1997) — A Neural Substrate of Prediction and Reward (Science).
  Science/DOI: Science | PubMed: PubMed

• Berridge & Robinson (1998) — What is the role of dopamine in reward…? (Brain Research Reviews).
  ScienceDirect: ScienceDirect | PubMed: PubMed

Executive function / media multitasking

• Ophir, Nass & Wagner (2009) — Cognitive control in media multitaskers (PNAS).
  PNAS/DOI: PNAS | PubMed: PubMed

• Commentary piece you can cite for framing: Lin (2009) — Breadth-biased vs focused cognitive control… (PNAS). PNAS

Self-interruptions / task non-resumption

• Adler & Benbunan-Fich (2013) — Self-interruptions in discretionary multitasking (Computers in Human Behavior).
  ScienceDirect landing page: ScienceDirect | Direct PDF (interruptions.net): interruptions.net

• Baethge & Rigotti (2013) — Interruptions to workflow… (Work & Stress).
  Taylor & Francis/DOI: Taylor & Francis Online | PDF (interruptions.net): interruptions.net

Social-media / “screen time” harm debate

• Orben (2020) — The Sisyphean Cycle of Technology Panics (Perspectives on Psychological Science).
  PubMed: PubMed | Publisher/DOI: SAGE Journals | Cambridge repo: Cambridge Repository

• Twenge et al. (2018) — Increases in Depressive Symptoms… and Links to Increased New Media Screen Time (Clinical Psychological Science).
  SAGE/DOI: SAGE Journals

Essay / book precedents

• Nicholas Carr — The Shallows (publisher page): W. W. Norton & Company | Author page: Nicholas Carr

• Cal Newport — “Some Notes on Deep Working” (essay/blog): Cal Newport | Book page: Cal Newport

• Adam Alter — Irresistible (author page): Adam Alter | Penguin page: Penguin

• Tristan Harris — US Senate testimony excerpt (includes “race to the bottom of the brain stem”, infinite scroll, pull-to-refresh): commerce.senate.gov

• Craig Mod — “How I Got My Attention Back”: Craig Mod

The common narrative pre-2025 has been that Europe lacks in innovation, that Brussels only cares about regulation and bureaucracy. However for every monochrome narrative on the internet there seems to be a powerful swinging back of the pendulum.

Indeed these days, there is a lot to be optimistic about (Europhoria?). Reforms at the EU level are slower than elsewhere and there are reasons for that, but I believe there is slow and steady change and progress.

Founders and VCs have moved forth an entire European movement.
EU-Inc was not a think-tank idea, nor something pushed by commissioners. It emerged organically from the community: operators, investors, repeat founders, and builders who were tired of watching Europe remain a passive fragmented consumer of foreign tech. Those actors didn’t wait for the institutions. They started acting as a unified economic block long before politicians did.

As much as Big Tech is what made Trump (Cambridge Analytica, etc), the emerging startup ecosystem in Europe is what is quietly forging the next iteration of the EU.

After all, isn’t the EU a startup of sorts? A political “project”.

The more European companies scale across borders, hire across borders, and raise capital across borders, the more integrated the continent becomes in practice rather than in theory. That economic and cultural imprint creates a form of cohesion that no directive or treaty alone could produce.

This intertwining of industry and integration will have a lasting effect.
Every wave of global instability, US–China rivalry, Russia’s aggression, the fragmentation of supply chains, pushes Europe’s further toward cohesion, not away from it, despite what adversary autocrats would want.

Think of all the EU-level reforms simply enacted this year. It represents seismic change.

The lesson of the last decade is that fragmentation is a luxury Europe can no longer afford. Even far-right actors will slowly start realizing that having a separatist national movement is stupidly unstrategic. Fragmented markets, fragmented army, fragmented foreign policy. Highway to geopolitical irrelevance.

And while most political actors and industry incumbents are still trapped in national narratives, tech founders are the only group consistently thinking at the scale the moment demands. Even Bernard Arnault has resisted certain federalizing reforms, not out of principle but because his industrial architecture is still anchored in a nation-state model. Founders, by contrast, have already moved on.

What is emerging is a blueprint for how tech can influence the EU, and how the EU, in turn, can support a generation of companies that operate at a European, not national, scale.

What this likely mean for the ecosystem.

The long-term trajectory is straightforward: technology will permeate and eventually govern every major industry in Europe.

Energy, food systems, mobility, health, finance, manufacturing, none of these sectors will remain “non-tech” for long. The shift won’t happen overnight, but the direction is locked in. Once digital infrastructure becomes the dominant layer across sectors, the actors who understand products, systems, and scaling will naturally shape the rules, the standards, and the culture that these industries operate under.

This is why the next decade matters. The US needed almost twenty years for its startup economy to evolve from niche subculture to the dominant force in national industry. Europe will follow a similar curve, but likely at higher speed, because it can leapfrog.

The playbooks, organizational patterns, and scaling models already exist. Europe can compress that evolution from twenty years into ten or fifteen simply by learning from the failures, excesses, and regulatory gaps that the US and China exposed.

And this is the part that most people underestimate: it’s not just about the technology. It’s about organisational design, scale dynamics, operating models, and culture. Picture a new business dynasty taking over as millenials take the reigns over from boomers and gen x.
A company that ships weekly will inevitably outperform an institution that ships annually. A team trained in product thinking will run circles around a civil-service hierarchy optimized for paperwork rather than iteration.

This creates an opening for founders:
build now, operate with the European scale in mind from day one, and the dividends will compound over the next 10–20 years. The companies that adopt this mindset early will not only define their industries, they will define how Europe itself operates.

What still needs to change.

For Europe to realize this trajectory, one shift is non-negotiable: we need to change our relationship to risk.

Right now, the entire ecosystem, from founders to investors to civil servants, is still shaped by a cultural bias towards risk-aversion. European VCs tolerate less downside volatility. Governments design programs to avoid failure rather than to maximise breakthroughs. Even socially, entrepreneurship is often framed as an exception instead of a norm.

If Europe wants to compete at the frontier, the risk profile of its venture ecosystem needs to bleed into the way government, institutions and everyday culture operate. That doesn’t mean copying the excesses of Silicon Valley. It means designing a system where taking calculated risks is also on the table, where early failure is not reputationally fatal, and where capital, regulation, and public programs accept that uncertainty is a feature of innovation, not a flaw.

Europe already has the ingredients: deep industrial expertise, strong human capital, and a continental market. When the cultural pivot happens, when the average policymaker, investor, and citizen becomes comfortable with higher risk bets, the continent will stop being a follower and start shaping the frontier.

🇪🇺📢Why does Europe need a sovereign media tech stack?

Every day, Europeans exist inside the American digital empire.

The moment we go online, we operate on platforms, protocols, and infrastructure built and governed in the United States. In a world of growing geopolitical fragmentation and fragile alliances, this dependence is becoming increasingly absurd.

For years, I’ve believed that the architecture of social media and post-internet media is one of the defining issues of our time. It makes little sense that a handful of companies control the global flow of information for billions of people and moreover optimized for engagement.

Yet when we discuss “tech sovereignty” in 2025, we focus almost entirely on servers, LLMs, chips, and the “AI stack”. We rarely talk about the foundational media layer that undermines our cultural systems.

Social platforms are part of the AI stack, arguably the most critical part.

Creating more or better content companies won’t fix this imbalance. As long as the platform and gatekeeper sit outside Europe, creators and institutions remain dependent. Traditional media or content-first strategies cannot solve the structural problem unless the underlying platform architecture is sovereign.

As we rebuild it domestically, we can build something infinitely better than the current web 2.0 Silicon Valley architecture.

Every day, Europeans exist inside the American digital empire.

⚡️🏗️What are the risks of the current setup?

1. State sovereignty

A state should help decide what information architecture it wants within its society, and this should be stirred by democratic elections. What is the point of electing our leaders if the President in another country decides what information we see?

Europe’s major political conversations run on platforms whose rules, ranking systems, and escalation policies are defined abroad.

2. Individual sovereignty

Users should be able to choose the algorithms that shape their feeds, and understand what goal each algorithm serves, whether engagement, learning, discovery, or networking.

Today, most of this logic is opaque and imposed without user agency. When individuals cannot select or understand the systems shaping their attention, their autonomy is compromised, and they are more easily influenced towards an economic or political outcome.

3. Cognitive preservation

If platforms optimize purely for engagement, the outcome is predictable: low-quality content rises, attention spans collapse, and “brainrot” becomes the cultural norm. We deplete the collective intellectual capital.

A consumer-driven media layer produces a population that is less informed and less capable of high-agency decision-making. No society can maintain cognitive resilience under an engagement-maximising architecture.

4. Security risks

Europe’s communication layer depends on infrastructures that can be altered, restricted, or weaponised during political or geopolitical crises.

Sudden moderation freezes, algorithmic shifts, or coordinated misinformation campaigns can cascade across the continent without European control. This dependence leaves Europe exposed at moments where stability matters most.

5. Cultural risks

European cultural expression is filtered through algorithms designed around U.S. engagement norms. Over time, this influences what becomes visible, credible, or culturally “central,” slowly reshaping European identity.

Cultural autonomy erodes when another region’s optimization logic determines what gets amplified.

6. Centralisation risks

Today, platforms, algorithms, and content creators operate as a single fused power block. This concentration allows a few companies to control distribution, ranking, and economic outcomes for the entire media ecosystem.

A healthier architecture would separate these layers: platforms, algos, and creators, just as anti-monopoly frameworks separate utilities, distribution, and production.

⏱️💥Why now? Factors that make it a good time to build it.

1. Geopolitical fragmentation is accelerating

Global alliances are becoming more fragile, and Europe is increasingly expected to stand on its own.

A sovereign digital ecosystem has moved from a long-term aspiration to an immediate strategic priority, especially as information influence becomes a core component of geopolitical power.

2. LLMs create a once-in-a-generation architectural reset

Large language models represent a quantum leap in how information is organised, ranked, and consumed.

The next media layer will be built around AI-native interfaces, and this window is still fluid. Europe has a rare chance to shape the new architecture before it solidifies around non-European defaults.

3. Europe is building the capacity to defend digital sovereignty

With Europe investing in collective defense and a more unified security posture, it is gaining the leverage needed to assert autonomy in the digital sphere.

A sovereign media stack becomes more feasible when Europe can push back against external pressure rather than relying on the goodwill of foreign actors.

4. Momentum for domestic solutions is stronger than ever

Across the continent, startups, policymakers, and institutions are aligned behind the need for European infrastructure in AI, cloud, chips, and media.

The conditions for local innovation, capital, regulation, and political will, are finally converging, creating the perfect moment to scale homegrown platforms.

Subscribe now

👁️🏗️What could this new media tech ecosystem look like?

1. Platforms, algorithms, and creators become separate layers

Instead of the Meta-style monolith, each component: content creation, distribution, ranking, and consumption, would exist as independent companies.

This modularity makes the ecosystem more open, more competitive, and easier to regulate, while AI dramatically reduces the cost of building each layer.

2. AI agents act as personal information curators

AI assistants would serve as your investigators and filters, searching the web, analyzing sources, and curating information based on your goals. Content would be queried through APIs rather than locked inside proprietary feeds, enabling a healthier and more transparent flow of information.

3. Users become the primary economic stakeholders

A paid or hybrid model ensures the platform serves the user rather than advertisers. When revenue depends on user satisfaction instead of attention extraction, the system naturally optimizes for quality, relevance, and wellbeing.

4. Strict identification and labelling of AI-generated content

AI output would be flagged by standardized rules, reducing manipulation and making the information environment more trustworthy. This helps preserve human agency in a landscape increasingly populated by synthetic media.

5. Algorithms are transparent and user-controlled

Users should be able to choose which algorithm runs their feed, whether the goal is learning, discovery, networking, news, or entertainment. A marketplace for algorithms would make optimisation explicit and put control back in the hands of citizens.

6. APIs create cleaner separations between data and platforms

APIs would become a default monetisation model, allowing data, ranking systems, and interfaces to remain independent. This reduces lock-in and prevents any one company from owning the entire media pipeline.

7. More private digital rooms replace the global town square

Smaller, interoperable communities enable safer discussion, richer relationships, and healthier discourse. This shifts attention away from mass-scale engagement and toward meaningful interaction.

8. Interoperability rules keep the ecosystem open

Platforms must be able to communicate with each other, making it easier for new entrants to build on top of existing infrastructure. This limits monopolies and encourages innovation across the entire stack.

9. Domestic content receives structural support

Prioritizing European cultural and informational output reduces foreign influence, bot farms and strengthens local creators. It helps cultivate a media layer that reflects European values, languages, and perspectives.

🤝👉What can the Europe do to foster it?

These are some incentives we should try and build as an ecosystem:

1. Strongly favor domestic platforms through protectionist mechanisms

Europe can shape incentives so that homegrown media technologies thrive, especially in early stages. Preferential treatment, procurement pathways, and regulatory fast lanes can give European platforms the runway they need to compete.

2. Encourage non-advertising business models

Tax benefits or reduced compliance burdens could reward companies that rely on subscriptions, micropayments, or user-owned models. This would shift the economic logic away from attention extraction and toward user value creation.

3. Apply antitrust measures to mature platforms

Once a platform reaches a certain scale, European regulators could require structural separation of layers, content hosting, ranking algorithms, advertising networks, mirroring the modular ecosystem proposed above.

This reduces concentration and encourages healthier competition.

4. Establish interoperability incentives

Common technical standards can help new entrants plug into the ecosystem more easily. Whether through direct mandates or incentives, interoperability would ensure that European platforms can grow without being blocked by incumbent gatekeepers.

5. Support creators who choose European platforms

Creators drive the cultural layer. Tax benefits, grants, creator funds, or preferential distribution partnerships could encourage them to build their audiences within European ecosystems rather than defaulting to foreign ones.

The societies that design media infrastructure to elevate education, agency, and critical thinking will hold the strategic advantage in the information age. Europe should treat cognitive resilience as part of its digital sovereignty, not an afterthought.

Share this post if you think this issue deserves more attention!

For most of the last few thousand years, our sense of morality didn’t come from within. It was imposed from the outside, through shared rules, religious codes, and the threat of punishment or promise of reward. A sort of outsourcing of our own intuition, because we aren’t trustworthy beings.

Whether it began with tribal instincts, religion, or Enlightenment thinking, morality served one central purpose: to align people around a shared sense of what is right and wrong. It created order by teaching children how to fit into society’s mold.

But today, that shared moral compass is breaking down.

Why is that?

We no longer live in a world with one dominant narrative. Instead, we inhabit fragmented realities, each person consuming a different feed, shaped by algorithms and cultural bubbles.

This didn’t start yesterday. The collapse of grand narratives, like religion, nationalism, and progress has been underway for over a century. But the internet has supercharged it, giving everyone a personalized worldview.

As a result, modern societies can no longer function as moral monoliths. There is no longer a clear consensus on what’s “common sense”.

Subscribe now

Signals of the shift

We’re not just theorizing here, this shift is already unfolding around us. You can spot its early signs in culture, technology, politics, and behavior. Below are some of the most telling signals that the old moral order is eroding and something new, messier, but more authentic is trying to emerge.

Cancel culture is burning out

The collective appetite for public shaming and moral outrage is declining. What once felt like justice now feels like noise, often lacking nuance, due process, or long-term impact.

Universal values don’t feel universal

Documents like the UN Declaration of Human Rights are increasingly seen as Western in origin and scope, reflecting only a subset of cultural worldviews. As global power shifts and pluralism expands, attempts to define a global moral baseline are viewed with skepticism, especially in the Global South.

AI ethics expose our divisions

When we try to encode ethics into AI systems, we hit a wall: whose ethics? What seems fair or just in one culture may be deeply problematic in another.

The moral left is losing ground

Progressive politics often lean on appeals to justice, fairness, and virtue, these arguments now struggle to cut through. In a world saturated with competing truths, morality has less persuasive power than value creation, power, or risk-aversion (read fear). This partly explains the rightward shift in many Western electorates.

Shaming is self-defeating

Shame used to work when society had shared values, but in today’s fractured landscape, it often backfires. People double down, dig in, or seek out like-minded echo chambers. The result is polarization, not progress.

Spirituality is shifting East

There is growing interest in Eastern frameworks that emphasize embodiment, presence, and personal awakening over rule-based morality. Practices like meditation, breath work, or somatic healing prioritize inner alignment rather than external obedience. This suggests a turn from moral authority toward felt experience.

Authority is questioned

Institutional trust is at an all-time low, from governments to science to media. Traditional moral arbiters are seen as compromised, biased, or captured by interests. When people stop believing in referees, they stop playing by the rules.

Conspiracies fill the vacuum

In the absence of trusted narratives, people create their own. Conspiracy theories often emerge not from stupidity, but from moral hunger, the desire for good vs. evil stories in a world that no longer offers clear ones. It’s a re-moralization of chaos.

OK - so now what?

“The old world is dying, and the new world struggles to be born; now is the time of monsters.” — Antonio Gramsci

If the old morality has collapsed, what replaces it?

In another piece, I argue we are exiting the Cognitive Era, a period defined by abstraction, rationality, and external authority structures. We built systems where morality was processed like software: rules, logic, categories.

But this requires something we’ve long neglected: reconnection to our own bodies, emotions, and instincts. Many people today are so somatically detached that they no longer know how to discern what’s good or bad outside of social approval.

But that model is breaking down. And here, I want to extend the argument: morality itself is evolving. We’re shifting from consuming morality, through religion, media, schoolbooks, to experiencing it. In the emerging paradigm, morality isn’t something we’re told. It’s something we feel, through intuition, embodiment, and lived experience.

For some, that sounds terrifying. Unstable. Even anarchic.

But take a breath. This shift won’t happen overnight. It will be phased, messy and uneven, and many will cling to old frameworks for good reason, they still offer structure in a disorienting world. Traditional morality still has a role to play, especially for those who aren’t ready to cross into the new paradigm. It gives people something to hold onto. And that's OK, for now.

Over time, though, it will recede.

One day, we may look back and wonder how we ever thought rule-based, cognitive morality was the pinnacle of ethics. It will seem mechanical, rigid, and disconnected, a relic of a time when we trusted thought more than felt sense, compliance more than consciousness.

The task ahead isn’t to rip morality out at the roots. It’s to push for an evolution, honoring what it gave us, while allowing something more organic and embodied to grow in its place.

Share

In 1994, while working on her Bedtime Stories album, Madonna reached out to Björk, then fresh off her debut solo release. Though initially reluctant, Björk penned the lyrics to what would become one of Madonna’s most experimental tracks: Bedtime Story

It begins with a haunting line:
“Today is the last day that I'm using words. They've gone out, lost their meaning. Don't function anymore.”

Ironically in the video she exists in some kind of AI simulation, and the logo of the company looks a whole lot like NVIDIA. I will leave it there and let the conspiracies spread!

The point is since I first heard the song, I have wondered: will there be a stage in human evolution where we will evolve beyond spoken language?

Is this just a step in our evolution? Before we exit the cognitive into the intuitive era? (read my previous article for context there)

Subscribe now

A step change

On November 30th 2022, ChatGPT was released to the public.

It marked a turning point: suddenly, all software could tap into the vast global database of human language. AI systems could not only understand this content, but also contribute to it.

This was partially the case before. There were chatbots, however the quality and distribution was no match. The origins of computing lie in codebreaking, Alan Turing's work on decrypting Nazi communications. Since then, computers have evolved into tools for interpreting, and shaping language at scale.

So yes, 3 years ago, all linguistic content became accessible to software systems given they found their way to it. This has two core implications:

1. Read access: All content anywhere on the web can be parsed and interpreted at scale by AI systems at the benefit of those who own them

   1. This makes our systems less resilient as it creates a higher potential for hyper-centralization. Think dictatorships where powerful institutions can predict your next move better than you can.

   2. It also creates an opportunity for chaotic decentralization where non-elected groups of people can acquire power by parsing compromising data of other groups

   3. It generally makes privacy more compromised in all directions

2. Write access: AI created content can be pumped to the web, and will soon no longer be differentiable from human content

   1. This makes any system or society vulnerable to external or internal malicious content

   2. It compromises the general quality of the content and information we are exposed to - which in turn compromises our understanding of the truth

This screams for an alternative human communication system that AI cannot observe, interpret, or replicate.

Don’t get me wrong, I believe AI can have a tremendously positive impact on the world, however regardless of where we stand, we need a control group and a variant. Not all our eggs in the AI basket. Evolution forces diversity. If a new intelligence is going to arrive among us, we need to create new forms of intelligence within the human ecosystem as well.

Now the question is: how do we create content that cannot be understood by AI?

I don’t mean this as an encryption problem, or at least not technological encryption. I am thinking about a literal human language that AI cannot understand.

Let’s take a step back.

Share

What is a language?

Definition: Language is a system for transmitting information from one individual to another.

Scope: The scope of a language encompasses all the concepts recognized within a given culture. There is a direct correlation between the breadth of a culture and the range of its language.

Channels: Most languages rely primarily on sound. Within spoken language, both factual meaning and emotional nuance are encoded. In some cultures, such as Arab or Italian, body language also plays a significant communicative role, blending gesture with speech to enrich expression.

What if we had a language beyond spoken word, that AI simply couldn’t decipher because it would require a human body to understand?

Now, when thinking about this I realized, when we speak to each other, there already is nuance that AI misses out on. The tone, the rhythm, the timing, etc. But it does not contain an important enough part of the intel for AI to stumble on the meaning of what is said.

The question is how do we lean in much further into the portion of language that is body-, tone- and rhythm-based? Think of when someone tells you something and it gives you the shivers, or there is just an unspeakable sense of dread.

Human speech is full of subtle signals, tone, rhythm, timing, that convey emotion and intent. AI can approximate these patterns, but they don’t fundamentally alter the meaning in ways that confuse machines. The core message remains decipherable.

The quest for new languages

The quest for people to speak in ways that is undetectable by AI systems is likely already underway.

I don’t know what the outcome will look like, but I know it will happen.

There are many questions I have not yet answered I want to leave you with:

• Will we able to communicate in this language using technology? (text, Facetime, etc)

• Where will this language emerge from?

• How many dialects will there be?

• How will these languages spread if not through technology?

• Will these need to keep evolving as AI systems will be able to decrypt them with an onset delay?

• Will speaking this language remain the privilege of the few in-groups?

I know that if you are reading this, you will think about it. And the very fact that this seed is planted, combined with our collective creativity will lead to great things.

Share

If you think about these topics and find it interesting, reach out!