Through the summer, I spent an hour every day walking around my neighborhood talking to myself like I was insane: headphones plugged in, waving my hands around like I was having a conversation with a real person, the whole bit… but it was just me talking to my AI agents and working through different projects I had on the go.

It’s a surprisingly effective way to get things done.

There was an evening in August where I went out and worked through:

• a full strategy deck for a startup I advise

• an investment memo on why people should buy first edition Harry Potter books

• a research report on energy and data centers for personal studying

• and mock ups for a few memes I’ll use in another essay

What hit me when I was out on one of these LLM walks is this mode of working is actually a phase shift in how we think as humans. In my mini model of the world there have been 3 eras of human cognition, and it seems like this year we entered a fourth, which I realize might sound like an hand wavy line of thinking so let’s speed run this and you can decide whether it’s convincing.

Here’s how I think about it…

First: humans had Hunter Gather Brain. People lived in small groups. They had voice-first societies and oral traditions (and so limited collective memory). In this era, the ear was privileged.

Next: humans leveled up and unlocked Phonetic Alphabet Brain. People lived in larger contexts, wrote things down, and information and ideas scaled on papyrus, cow hide, and clay tablets. In this era, the eye was now privileged.

>While the printing press was invented a couple thousand years later, I think it makes sense to understand that technology less as its own thing and more as a turbo-charged extension of the phonetic alphabet era.

Third: we got Computer Brain. People were able to outsource aspects of their thinking to a digital machine for the first time (and not like an abacus or whatever). This is the Steve Jobs idea that computers were/are a bicycle for the mind. We were given the ability to access and create information in an augmented manner. First locally, then at a global scale with the internet.

>Phones are an accelerant of this. We collapsed extreme processing power into a small object that we all carry with us, and while that sort of turned us into proto-cyborgs it really is just an intense version of computer brain.

Now, what makes this interesting to think about is it seems like AI teleported us into a fourth era of cognition that merges all past eras while creating something new as well. We keep our upgrades, but now we get superintelligence to help us with our tasks, and we can outsource the creation and transformation of information to external entities (AI agents) that go off on quests on our behalf.

Let’s call that Orchestration Brain.

This is insanely cool in part because we now have superhuman leverage for our thinking and acting in the world at all times. A demigod in our pockets. I can go for a walk and work for a few hours with both voice and visual feedback and create what would have been a week’s worth of work a couple years ago.

But working (and thinking) like this has a cost too. It’s a cost that’s tricky to discern at first because replacing screen-first grunt work with voice-first LLM work is useful, but I know I lose something in the process.

There’s this concept in The Expanse, which is a sci-fi book series (and TV show) where people who live on the moon or the asteroid belt grow up under weaker gravity and so their bones adapt accordingly. What that means is when these people come to Earth’s heavier gravity their bodies can’t handle the pressure so it’s excruciating for them to even walk or move around.

I worry that using LLMs is sort of like living in low gravity but for the mind: easy at first, but you only realize after returning to “real” work that you’re no longer able to do it well (or at all). So the answer is to panic, panic, panic and delete everything, and log off forever. No, lol.

I don’t know what the “answer” is to this, but I do want to frame the problem because there are power ups related to using agents that are very cool, but there are definitely costs too, and everyone who cares about preserving their ability to think will have to find their own way of threading that needle. It’s a bit goofy, but I think the ideal model here is Aladdin because he has access to his genie’s powers but has to use them sparingly because he only has a few wishes. And I think that’s the best way to use AI as well, even if we have to impose our own constraints.

The way I’m trying to live this out in my life is just prioritizing offline time to offset my LLM use. I want to sit and get my ass kicked by a blank page while writing, or re-read a hard paragraph in a book and do the work to preserve my brain. Then I want to choose my spots each day, lever up, and use this technology to make magic happen. So I guess the answer is not being completely anti-AI and it’s not being completely pro-AI either, but a secret third thing: being (very) intentional.

Thank you to Mark for feedback and review.

The first company I ever invested in imploded so quickly it was like I had a reverse Midas touch of instant death. This was back in 2022: someone asked me if I wanted to invest $5K in a crypto startup that was raising money at a nine-figure valuation, and I said yes, both because I’d never been given that sort of opportunity before and because it made me feel cool.

Fast forward a few months later… the bubble at that time popped, and I learned a quick lesson in how angel investing is mostly a fancy way to lose money.

The founders weren’t scammers per se – I mean, low bar – but their product had no moat, the market segment they were positioned in was shrinking, and they were too lazy to do much about anything. The token is down 99.9% since then, and my little stake (which is still vesting) is worth like 84 cents total. Makes me laugh whenever I think about it.

Welcome to the big leagues, kiddo.

I took a few months to sit on that first L and think about whether I wanted to keep throwing money at startups like this or not because I had opportunities to make investments, but I felt a lot of imposter syndrome.

What I ultimately decided to do was: keep investing, but think about it (mostly) as an educational exercise. This was inspired by some old Tim Ferriss content where he pointed out that people who are early in their careers are more likely to learn useful lessons by being in the trenches with a few real businesses than reading case studies all day in an MBA program.

So that’s what I did. I had some funds saved for going to grad school and instead of doing the academic thing I decided to use that money to invest in startups.

It’s been four years since then, and I still feel very green to early-stage investing, but after making 20 small investments (across crypto, AI, gaming, robotics, and healthcare) it seems like a good time to take stock of what happened along the way. And that’s the plan for this post: it’s a speed run through everything I’ve learned about startups and investing in one shot so get ready.

Positioning:

There are different ways to play the angel investing game and different ways to get access to investment opportunities.

Some people don’t have to do any work at all and they just receive inbound offers to invest in new companies because they have specific knowledge in a domain, so founders actively want their help and advice. Think: an AI researcher at one of the big labs getting pitched to invest in a new AI startup.

Some people win deals because of sizing: Rich Person X can shove a chunk of money into a startup at a moment’s notice and that can be attractive when all a founder needs is cash.

Other folks win deals because of personal connections to a founder. Maybe they’re family, former coworkers, college drinking buddies, whatever. They get to throw money in just because.

Other times it’s about visibility: there might be an influencer or another founder that’s active in the same niche as a new company, and having those folks invest can lend the startup in question some credibility. The extreme version of this is straight up famous people dropping angel cheques. (See venture funds run by/around Kim Kardashian, Jake Paul, and Ashton Kutcher).

TLDR: There aren’t any rules. You just need some money and an angle.

Now, I’m obviously not famous (yet), and I knew upfront that I was never going to win a deal through money firepower because I only had 20 shots to invest a $5K cheque, which is tiny sizing in this arena, so my whole strategy was to find cool startups early – ideally before they raised any money at all – then convince the founders to let me be a small part of their universes. And my angle was to pitch myself as a professional friend, like someone who will walk with you through the chaos of startup land day-to-day and help solve problems. That, and I give passable “storytelling” advice, and I have good sticker packs on Telegram to lighten the mood. So the trade offer was: if you let me invest in your company, I’ll help you out in those ways.

20 founders said yes.

Adding Value:

After cheques get written the energy shifts away from selling yourself and your capital as this very special and differentiated thing toward trying to help these companies win for real. In theory, whether it’s as a venture fund or as an angel, this sort of investing is supposed to be a services business where you offer a startup a mix of capital, connectivity, and advice. (This is what people mean when they say they’re ADDING VALUE.)

On this trifecta… we’ve talked about capital. I don’t have much of that, plus once a team gets my cheque that bit is done. When it comes to advice, my view is most business advice (mine included) is nonsense beyond the basic stuff because almost nothing generalizes well across different categories, and the world is changing so fast right now that a lot of previously good advice is decaying in relevance quickly. I do think there’s some value in pattern recognition, so when I’ve seen founders fail in particular ways I think flagging that for other founders who are facing similar situations can be useful, but I’ve largely focused my Value Adding Energy on connectivity.

What does that actually mean?

In practice, this has looked like: helping teams raise money, helping them hire good people, helping them get beta users, funneling relevant information to the founder(s), going to their events, and mentoring junior employees in media and go-to-market functions. I mostly live in the squishy, relational aspects of business. My hope is that I create some value for each team that I invest in, but if I’m being real I’m mostly a cheerleader. It’s just fun that these founders let me in to see the mess (and the magic) of trying to make something out of nothing.

Track Record:

One of the tricky aspects of investing in startups is it’s difficult to tell whether someone is any good at the game over a short time frame. It takes a long time to build a successful company, and most venture funds are 10-year funds to reflect that because you have to let the founders cook for while. Also, even in scenarios where an investor does have a good investment outcome it can be unclear whether that happened more because of luck or skill.

I say all this because for me personally I think the jury is still out on my investment career. I’m not going to sit here and tell you I’m good at this yet. It hasn’t been that long, and I haven’t had that many reps, so I’ll have to report back in 2030, but here’s where I find myself at the time of writing with the 20 deals I’ve done:

-10 companies are alive and have raised additional capital at higher valuations

-5 companies tried stuff, decided to wind down, and gave partial funds back

-4 companies went to zero and closed up shop completely

-1 company was modestly acquired

On paper this combines into a mini portfolio that’s up 2.72x over 4 years on the original money I invested. Though take that with a grain of salt because I’ve never sold or realized any gains from these positions and this stuff a) isn’t liquid and b) can always go to zero.

I go back and forth when I think about my performance so far. On one hand, it feels like a pretty average outcome which is lame to me because I’m competitive and I wish I was up infinitely more. There’s also no clear turbo winner in this batch of investments (yet). On the other hand, it’s a decent start, and these numbers could look very different in a couple years so we’ll see.

Mistakes:

At this point, I’ve made many (most?) of the mistakes that an investor can make:

-I was dumb, did the wrong deals, and those went to $0

-I was dumb again and didn’t do the right deals, or when I did… I didn’t size those deals correctly i.e. I kept my cheque size basically the same even though I had more conviction in some companies versus others

-I got trigger happy and funded too many things over too short of a timeframe out of fomo, and I should have been more patient and waited for better opportunities

-I invested in founders simply because they were similar to me in temperament, not because they were well suited for their startup

-I got finessed once by a friend who was just a good salesman and not a credible founder

-I worked really insane hours for no pay to help a company over 2 years which then collapsed and went to $0

-I invested in a team that I never met and they ghosted me

On and on. The list of my fumbles could be a full blog post on it’s own. This stuff is hard.

Highlights:

One of the best parts of investing is getting to learn from founders who are obsessed with their Thing. It’s insanely cool just getting to see the world through the eyes of a hyper-competent founder (no matter what industry they’re working in), and I’ve received some of the best education of my life just by being in proximity to these people.

Learning like this scales up over time as well. The more startups you invest in, the more you can cross reference how different founders engage with similar problems: How are these people adapting to AI? How do they retain talented employees? Are they managing their treasuries differently? On and on.

When I think about the moments that have felt the most rewarding in this line of work it’s not the learning though. It’s also actually not the times when a company wins in an obvious way like raising more money at a higher valuation. The real special moments are the messy ones where a founder invites you in, shares a problem, and then you get to work through it together and find a way forward. The people I advise figure out their own problems 99% of the time and I just hang out on the bench, but it’s definitely cool to check in the game once in a while and drop a little assist.

Chaos Mode:

Something that’s surprised me about this whole investment guy arc has been just how much business fuckery I’ve gotten to watch up close. So far, I’ve seen:

-2 different board coups where the founders/CEOs were cut and replaced

-An actual poison pill scenario where a company got recapitalized and my equity was diluted to 0

-Cyber attacks destroying the brand value of a company to the point where they had to wind down

-Cofounder breakups

-Teams pivoting into extremely different industries and then failing

-IP issues, noncompete issues, and a bunch of lawsuits

-Founders who turned VC funding into a lifestyle business so they could screw around while doing minimal work

-A team letting their core product die, then building an entirely new business (while using investor money) and then not including us in the new company’s cap table

All together, it’s been wild. Though seeing a bunch of different ways that companies can struggle has been valuable in the spirit of Tim Ferriss and investing-for-educational purposes, so in a sense it is what I signed up for.

Lessons:

Most of the anecdotes and observations that I’ve collected over my short investing career roll up into three lessons…

Lesson #1:

Startups are insanely hard. Most fail, and the ones that succeed have to get a lot of different things right, which makes picking companies to invest in a chaotic (and fascinating) betting game. It’s like betting on a poker hand pre-flop. You always have some information: in the case of poker you have your hole cards and in the case of a startup you can look at a founder, their product, the market they’re competing in, and so on. But both are games of incomplete information and that’s what makes them so captivating.

For betting on startups specifically, from TBPN captured the vibenomics of it well with this old tweet:

Lesson #2:

All the successful founders I know are unique, but they do have a few common traits.

First, they’re obsessed with their problem in a way that only they could be based on their life experiences, whether that’s fueled by passion or pain. I’ve always had the sense that the best founders would be working in their domain anyway, even if they didn’t have a company. They’re committed to their niche even in a world that’s constantly producing new shiny things to chase. Ben who is the founder of Cambria is a good example of this with gaming.

Second, these people have motion. They get stuff done and are good at reinventing themselves and their products. Business is a complex adaptive system where different companies and people compete to sell goods or services, and moving quickly across everything from hiring to narrative to product features and more is how you win (and continue to win) the game. Few startups move fast enough, and it becomes obvious when you work with a team that has the right pace that pace itself is everything. Andrew who is the founder of Awaken is a good example of this with tax software.

Third, the best founders hold ambiguity well. They persist when things aren’t clear. They can take the pain of waiting and failing and pivoting and being told they’re wrong and they’re not going to make it. They have conviction in their ideas and that helps them hold on through the ups and downs of startup life. Charli who is the founder of Web is a good example of this with community software.

Lesson #3:

Investing in startups is a genuine skill. This might seem obvious, but for a long time I thought venture was more-or-less a game of luck and timing. There absolutely is a lot of survivorship bias in the industry, but none of that changes the fact that the top tier people can consistently make good investment decisions that turn into massive wins for them and their firms.

How do they do that? It’s tricky to quantify. After watching some killer investors up close and from afar it does feel like more of an art than a science. There’s a scene in Gladiator 2 where Denzel Washington’s character – Macrinus – explains how he picks gladiators that I think articulates this point about venture well. He’s talking to the main character of the movie, who he bought to fight in the Colosseum on his behalf, and he says:

“It is an art, choosing gladiators. Some choose entertainers. Some choose brute force. I choose rage.”

And angel/venture investing is similar, especially at early stages when most investment decisions hinge on the founder or the team. Great investors all seem to have their own way to identify talent at that stage (or choose their fighters).

Getting Better

There are different ways to increase your likelihood of doing well as an angel. To some extent it’s about access, relationships, and network: are you seeing the right deals at the right times? To some extent it’s about breadth: are you writing enough cheques such that the math of venture investing can work itself out and you get a win that pays for all the other losses? To some extent it’s about decision making: are you making good investment decisions each time you see a deal?

I need to work on all of these areas, but the third one – when to bet and when not to – is the most interesting to me.

Here’s how I’m trying to get better at it:

1) I want to go slower, operate with less fomo, make fewer bets, and be comfortable missing stuff. I’ve thrown money away just because it seemed cool to have access to a person’s startup. I’ve anchored to recent deal flow and made investments because it seemed like Company Z was relatively stronger than Companies A through Y. But just because you see something doesn’t mean it’s good, and just because one thing you see is better than another thing, doesn’t mean it’s fundamentally good either. The reality is most deals are trash and you have to be paranoid about keeping your quality bar high.

2) I want to focus more on founder-market fit and weigh that more significantly in my decision making process. My working theory is founder-market fit is the most important variable in early-stage investing today (I’m assuming the market in question is at least growing.) Everything from founder pedigree and the current version of a product to the team and the timing of their raise and blah blah must all bow down to the question: is this person the right person to build something in this market for the next 5-10+ years?

3) I want to double down on investing in companies when I’m the first money committed. I don’t live in San Francisco or New York, and it’s difficult for me to compete against those networks, but the best outcomes I’ve had have happened away from the ruckus anyway i.e. helping founders before they’ve even raised – literally as they’re trying to figure out what to do – so I want to focus there because if I have an edge at all that’s where it is.

The Show Goes On

I originally started investing in startups to learn as much about business as I could as quickly as possible and the last four years have delivered on that.

On what comes next: I wish I had an insightful view to offer on the future of venture to wrap this up. I don’t. One thing I’m thinking about these days (along with everybody else) is to what extent AI changes the investing profession over the long term. I think LLMs can already do ~70% of an investor’s job reasonably well across admin, research, deal sourcing, and due diligence… but my sense is there will be a human in the loop for a long time because this is fundamentally a relationships business.

Finally, even though I’m out of the initial funds I had set aside for investing, my plan is to keep writing a few small cheques to a few special startups each year because I love this stuff. Investing is such a fascinating dance of trying to balance:

a) having humility before a complex future while…
b) expressing views on how that future could play out while…
c) trying to back real human people who will thrive regardless of conditions, all with the intention of…
d) trying to (in small ways) support or manifest a future that you want to see.

Few games are more fun.

A big thank you to Jon Moore, Smac, and Hype.eth for feedback + review.

In April, I gave a (very) short talk at Dorm Summit to a bunch of college students on how they might think about their careers through the lens of reputation and personal brand. It was mostly me telling a bunch of stories and sharing memes. If you’re young, or if you’re thinking about what to do with your life, or if you’re otherwise just interested in listening to me yap for 14 minutes on those topics, then you can check out the recording below.

Zack and the DormDAO team did a great job with the event, and you can learn more about them here as well.

Over the past month, I built a plugin for Claude to help with the process of writing essays. It’s called Riff, and it’s now on version 2.4.8. For folks who are unfamiliar, plugins are basically mini expansion packs that can be used across Anthropic’s different products to help Claude do specific tasks, and in this case, my plugin contains almost everything I know about writing, framed specifically to make writing with an LLM more intuitive through different steps.

I’ve been obsessed with making a tool like this because I think writing with AI can be empowering if done correctly, but most people struggle to use it well because they treat writing as a product to be created instead of a process to be supported. I don’t think good writing can be generated with a few prompts (even with the best models today) because it takes time and original thinking to write anything worth reading. That said, while AI systems can’t replace the squishy human inputs that illuminate good writing, they can meaningfully help with the writing process itself.

Riff specifically is built to be an extra layer of support for people who want to write essays, and the plugin is designed around two principles. The first is that the writer always leads. Every skill in the plugin presents outputs or options to help move the writing process along, then waits for the writer to decide what to implement. The second principle is that writing has different stages that each require different kinds of help, so each skill in here is built to focus intentionally on one phase of the writing process.

There are nine skills total. Most of them are designed to work together in a pipeline, though each one also works on its own if you just want help with something specific. I’ll walk through them quickly so you have a sense of what’s going on.

/sort takes unstructured material like notes, bullet points, or voice memos and distills it into a set of core points. It assesses the type of material you’re working with, consolidates overlapping ideas, and preserves your original language, phrasing, and references. Anything that doesn’t fit the main structure goes into a “junk drawer” so nothing important gets dropped.

/sequence takes sorted points and helps you figure out how to arrange them into an essay. Before proposing structures, it asks what your goals are with a piece of writing to clarify your intentions, then it proposes one to three structural options depending on what the material supports.

/compose writes a full personal essay from structured notes in your voice. The first time you use it, it asks for a writing sample so it can build a style profile that captures how you actually write. That profile is saved and reused across sessions. It also asks where the essay will be published and how long it should be (to calibrate tone and length).

/critique gives an honest assessment of a piece of writing. It opens with the single most important thing to fix, then works through remaining observations in descending order of severity. It evaluates originality, argument quality, fluff, tone, and vulnerability to criticism, focusing on the dimensions that are most relevant to your specific piece. It closes with three simulated social media reactions to show how the piece might land in public.

/revise takes a draft and a set of feedback and produces a stronger draft. It implements feedback point by point while preserving your voice and everything that was already working in the underlying essay. It can also recommend stepping back to use another skill if the piece isn’t ready for polishing yet.

/copyedit does a close read of a piece of writing to check for grammar issues, punctuation errors, and sentence rhythm. Each suggestion cites an existing section, states its proposed change to that section, and gives one sentence of reasoning for that change. It can implement whatever changes you like.

/title helps you find the right title for a piece of writing through analysis of the essay’s argument, imagery, and style. It generates title ideas across three categories: two words, three words, and longer (up to 10 words), then iterates based on what resonates with you.

/checkpoint lets you save, view, and restore versions of your essay as you work on it. You can browse saved versions anytime in the local folder that you’re working in, or restore an earlier draft if a revision went sideways.

/riff is for getting unstuck. Call on it at any point in the process and it’ll give you a quick, honest read on where things stand with a few practical observations and a nudge toward what to do next. The /riff skill also acts as a fallback when other skills aren’t triggered.

If you’d like to check out the plugin, experiment with it, or just learn more about what’s happening under the hood with the +9000 words of context across these skill files, then take a look at the project repo on GitHub here. It’s free and MIT licensed, and the README file there has installation instructions. If you find it useful, you can help me out by starring the repo as well. It helps more people find it. Also, if you use AI agents beyond Claude, the plugin is available on skills.sh here.

Riff is experimental. I update it often for my own use, so if something doesn’t work or if you have ideas for how to improve it, please feel free to hit me up. All feedback is welcome. If you download this locally for use in Claude Cowork, you can also edit each skill file individually to bend it in the direction of what works best for you, so really my version is just a starting point. No matter how you use this, I hope it makes your writing process a little easier and a little more a fun. Now go write something.

Thank you for reading my writing. I really appreciate your time and attention. If you’d like to get future essays in your inbox you can subscribe (for free) below.

Subscribe now

Everyone I know that works in tech is having some flavor of existential crisis these days. People are worried that advanced LLMs are going to replace all knowledge work, possibly even in the near future.

I think part of this anxiety is reasonable. The AI labs have released impressive updates over the last six months, and the industry is alive with what feels like exponential change, but despite all the hoopla (and very real innovation), what’s inconsistent to me is that I still find AI systems incredibly underwhelming when it comes to writing.

What I want to do with this essay is reason through why I think that’s the case and by extension wrestle with where I think human writers fit in a world of (increasingly) elite artificial intelligence.

To start, I want to be specific about what I mean by writing because it’s important: writing is thinking on paper. It’s an extension of someone’s mind made manifest in the world through words with the purpose of passing along a container of meaning from one person to another.

I realize that’s a broad starting point, so let’s go one level deeper and split writing into two sub-categories: informational writing and interpretive writing. Informational writing is about the transmission of facts. It’s the world of documentation, instruction manuals, and neutral reporting. Interpretive writing is about sharing a perspective. It’s the world of essays, opinions, and commentary.

I acknowledge that life is more complicated than a simple binary, and, yes, every piece of writing has some subjectivity to it with how information is presented, but for the sake of argument, let’s say that all writing has a dominant intent, which is one of those two categories.

LLMs are great at informational writing. They can create a memo from a voice call, summarize an article, draft a simple contract, and so on. That’s useful for work purposes, but it’s not what I’m talking about here. When I say LLMs are weak at writing, what I mean is they’re weak at interpretive writing. They do poorly when trying to write with a consistent subjective lens. They struggle to have an opinion.

One final clarifier before moving forward: let’s use essays as the sole example of interpretive writing so we can agree on a concrete form to keep in mind as we discuss these ideas; and while we’re here, I’ll give some background on what an essay actually is so we’re ultra clear.

The word essay comes from the French, “essai,” which means to try or to attempt. The connotation there is of wrestling with words to work through a problem. So, we can say that an essay is struggle in written form, and so far, the machines can’t write in that manner. There are two reasons why I think this is the case. Let’s consider both.

The first problem with LLMs writing essays is the entire process of prompting one of these systems to get an output misunderstands what the act of writing is.

People fundamentally can’t prompt good essays into existence because writing is not a top-down exercise of applying knowledge you have upfront and asking an LLM to create something. AI agents also can’t create good essays for the same reason. Even though their step-by-step reasoning is more complex and iterative than human prompting, a chain of thought is still trying to accomplish a predefined goal.

By contrast, real writing is bottom up. You don’t know what you want to say in advance. It’s a process of discovery where you start with a set of half-baked ideas and work with them in non-linear ways to find out what you really think.

There’s a Flannery O’Connor quote that speaks to this where she says, “I write because I don’t know what I think until I read what I say.” I’m the same way. I don’t think people can skip doing the work of wrestling with ideas because that’s the core chaotic energy that underpins good writing. It’s how you figure out what you believe and how you want to communicate it.

There’s also a more philosophical extension to this point which is that human brains function differently than LLMs. They’re squishier. There’s a famous book, Thinking Fast and Slow by Daniel Kahneman that has a helpful framework to explain this difference. He talks about two types of thinking. There’s System 1 thinking, which is fast, automatic, and driven by pattern recognition. Then there’s System 2 thinking, which is slow, methodical, and reflective.

Humans think in both modes. LLMs also “think” in both modes.

For System 1 thinking, a human makes snap judgments all the time about people they see, posts they read on the internet, or whatever else. Similarly, an LLM can take in a simple query and give us a one-shot output in response based on the model’s training.

For System 2 thinking, a human takes in ideas, considers them, and through conscious effort can produce new ideas. Similarly, reasoning models can think for extended periods of time, consider a problem, and produce novel outputs.

The difference – in both types of thinking – is that LLM “thought” or inference is mathematically deterministic where human thought is delightfully messy.

Once a model finishes its training, its weights are saved and its thinking function becomes static. This is represented as a set of numbers, which from that point on don’t change. When a model thinks, it takes an input and iteratively chucks that input into a vortex of math (its weights and operations) and that then produces an output. It’s a predictive exercise based on statistical associations, and it only feels alive because when words are generated there’s a pseudo-random sampling involved in that output process.

Now, when a person thinks, our manner of thinking is more like taking ideas and filtering them through a set of “weights” that are constantly in flux, namely our accumulated life experiences and how we’re constantly interpreting and reinterpreting them. The process is alive.

What happens is we take an input thought and refract it through this prism of our humanity, our memories, the places we’ve lived, the arguments we’ve had, the people we know, and infinite other bits and bops of life experience to see multiple, blended, possible outputs on the other side simultaneously. It’s like the Pink Floyd album artwork below where normal light is the input, the rainbow is the set of outputs, and the prism is someone’s brain.

So, both LLMs and human brains can think and compose writing. That’s cool. The issue is the way these machines produce words leads to sterile outputs. Like, it’s always obvious to me when someone shares writing they’ve prompted because the outputs tell on themselves. The reason for this is because LLMs lack the unique perspective of a human mind at work (or play).

There’s a visual analogy I like to use when comparing these two kinds of writing which is that a good essay is like a cathedral of ideas. It’s a work of craft, beauty, and the quirks of time and place. But an LLM attempt at an essay is like a glass skyscraper. It’s functional. It’s a building, sure. But if you looked at a picture of it, you’d have a difficult time telling whether it was located in Portland, Atlanta, or Denver because it has no soul.

The best counterpoint here is that I simply have a skill issue on my hands. Maybe I’m not clever enough, and if only I knew how to more skillfully orchestrate LLMs or swarms of editorial AI agents, then I’d be able to create works of magical, overflowing craftsmanship with them.

Maybe.

I do think people with specialized knowledge of writing can create better outputs with an LLM. Agents can get to slightly better outputs as well through iterative loops. But there’s an upper bound to how far you can get in terms of making an LLM compose something compelling because writing is a complex, adaptive process and these systems operate on a flattened representation of that.

My next point is that LLMs struggle to generate meaningful writing because they lack context. They simply can’t take in (or process) the staggering amount of information that’s involved in the act of a person having an opinion.

I recognize these models have proven that they have PhD-level intelligence in some areas. They can win math Olympiads. They’re trained on most of recorded history. There’s no question that their form of alien cognition is impressive, but it’s just not equivalent to the neural network of my brain or yours when it comes to writing.

Let me use a graphic to explain this view more clearly.

If I write an essay, there are three variables in that process: myself as the writer, my audience, and the essay topic. Also, each of these variables is constantly changing.

A person is dynamic. They’re always in motion. Everything from how I’m feeling and what I’m currently reading to random mood swings and mental biases shape how I write something.

An audience is dynamic in the sense that their relationship to reading is evolving. Attention spans change, content formats rise and fall in popularity, again everything is in motion. For example, people who are familiar with LLMs have noticed that they tend to produce a lot of em dashes in their outputs, which in turn makes writers wary of using them – they don’t want their work to be discounted as “just” generated by AI.

A topic is also dynamic to varying degrees depending on what we’re talking about. Some subjects like ancient history are more static, but generally the world is a fluid place.

To dial this up further, the relationships between these variables are changing as well! A writer shapes the expectations of their audience through what they write over time, and in turn a writer’s audience informs what that writer writes about when they show interest in one subject over another. Both the writer and their audience also have evolving views of a topic as time moves forward and new information emerges.

To tie this together: there’s a lot of social complexity that goes into writing an essay, and an LLM has limited visibility into most of that complexity, which limits its ability to write with the flair of a human or on behalf of one.

Further, this idea that LLMs lack context is more than a raw data problem, it’s a relationship problem. Human writers have a qualitatively different relationship to this information than LLMs do. I don’t have data about myself… I am me. I have an intuition about my audience. When I write about a topic, I don’t know about it, I care about it.

LLMs don’t operate on the same plane. They have a view, but not my view. They have some context, and maybe often too much context, but not the context that’s relevant to play this game well. And since an essay really is just a snapshot in time of a person’s brain as a function of all these variables this makes human writing a lot more defensible than some people suggest.

After chatting about these ideas with friends, I’ve heard a couple good counterarguments to my issues with LLMs and writing. The most convincing of these is that everything will be solved when AI systems get access to infinite (or drastically larger) context windows.

This line of thinking says that if LLMs can offer users, say, 500 million tokens as a context window, then our machine companions will be able to reference everything about us: all our past pictures, writing, social media history, health data, recorded memories through things like wearable glasses, and so on. Then they’ll synthesize all of this data to approximate our decision-making process, which in turn would let them write (and think) on our behalf.

I do think there’s some merit to this. What’s happening right now with AI agents that initiate their sessions with access to memory files, conversation history, active skills, and other reference data is definitely gesturing at this future. See below for my basic bitch graphic.

I will say that I think this stuff is extremely cool, and I do think we’re going to get closer to modeling human consciousness through the combination of agent architectures and larger context windows. But no matter how weirdly fascinating it is to conjure up an agent with specific characteristics, I remain unconvinced that these systems will be able to write on my behalf.

I have two disagreements with this supersized context window argument, both of which are more philosophical than technical.

For one, even if you get infinite context windows, LLMs have an issue with hierarchy of information. There’s no clarity around the weight of some information over other information and how that changes in different contexts. So, Claude might have access to my references, but it doesn’t know when and how to use them. It might know grammar and rules of composition, but it doesn’t know when I might paraphrase something versus quote it outright. And there are far too many exceptions to any rule to ever codify someone’s intuition.

Next, I just find the comparison between an LLM and a human mind misleading. We’ve talked about this a bit already. It’s a case where reasoning by analogy leads to the wrong conclusion because LLMs don’t behave the way human brains behave.

I do think LLMs are useful machines. I’ve used ~20,000 words of my writing as a reference point, written up rules around how an agent should make editorial decisions on my behalf, then with that context I’ve run ideas against the best models that exist today. It’s created some fun outputs. But in every case, something feels off.

In music terms, I think LLMs are like a cover band that’s trying to perform a song where they get aspects of it “correct,” but they’re never able to embody what makes the song special because they lack the essence that is embedded in the original musicians. I might have a great time listening to the music, but I’m not paying $400 for a ticket to see PISS live.

The AI maximalists will double down and make the case that, again, extra context is all you need to solve these issues, so one day an LLM will simply absorb mass amounts of your data and voila. Apparently, you can model everything about a human.

I disagree.

I’m not convinced we humans are even so legible to ourselves that we could articulate all the relevant context that an AI would require to mimic us.

My favorite example of this is the Greek concept of apophenia, which is the idea that humans perceive connections or patterns in random data. We don’t know why this happens, but this feature of our operating system is why we see different shapes in clouds, where, for example, you might see Bart Simpson and I see Pikachu. That creative expression is unique to each of us. How do you model that?

It’s a bit ironic, but our irrationality might be the most defensible asset we have as far as creative endeavours go. That bug is our best feature in a world of rapidly advancing AI systems because it makes us impossible to model.

put this well in a blog post last year where he said, “Lots of people worry that AI will replace human writers. But I know something the computer doesn’t know, which is what it feels like inside my head. There is no text, no .jpg, no .csv that contains this information, because it is ineffable. My job is to carve off a sliver of the ineffable, and to eff it.”

Okay, this has been long. There are other counterarguments to what I’ve laid out here. I just want address one more.

Some folks bring up coding and point out that it’s a complex, creative activity like writing, and the follow up question is: if LLMs have changed how coding is done to the point where it’s “solved” why can’t writing be solved as well?

I’m no authority on coding, but I do see that LLMs have changed that activity. My pushback there is that coding and writing ultimately have a different telos.

The way you evaluate writing is whether it resonated with an audience or not. It’s about curating an entertainment experience for another human. Coding is evaluated by correctness. It’s functional where writing is experiential and concrete where writing is fluffy.

There are for sure multiple ways to code a piece of software to accomplish a task, but there are more standards to follow than there are in writing, plus, as much as engineers care about the readability of code, the ultimate audience for it is a compiler not a human.

I think LLMs work well in coding because it’s a coherent system: an agent can create code, run tests, iterate, then verify whether it will execute or not. But with writing, that doesn’t apply because it’s not a computational exercise, it’s an interpretive one. As a result, LLM code is mostly coherent these days, and when it’s not it can be concretely edited. LLM writing is way less coherent.

An analogy that I like to think about for LLM writing outputs is they’re similar to the bar scene from Good Will Hunting where there’s a dweebie Harvard student who presents what appears to be a fancy argument about US history, but it turns out he’s borrowing someone else’s thinking, and Will goes, “Were you going to plagiarize the whole thing for us? Do you have any thoughts of your own on this matter?”

That’s what it’s like talking to ChatGPT. It’s more like the illusion of intelligence than actual intelligence.

As I wind this down, I want to say that I’m optimistic about AI. I see LLMs as a complement, not a substitute to our intelligence. They’re super useful tools, and I use them in my writing today for ideation and sometimes for editorial feedback.

That said, even in a world where we see significant improvements in AI systems, I find it unconvincing that they will replace human writers. There’s a special je ne sais quoi about being human and the way we hallucinate our own custom models of the world based on a lifetime of experiences that I don’t think any machine will replicate.

Now, could I be wrong? Possibly. Will GPT 33 be able to write perfect essays? We’ll see. Models have surprised me and I’m sure they will continue to.

But my view is good writing requires something that only humans have, which is the ability to struggle and to speak about that struggle from the point of view of being in the trenches of life. It is sense-making from one human to another, and in my mind as long as humans endure as a species, there will be an audience for that.

Until then, keep writing.

A massive thank you to Devin, Finn, Nuk, and Luffistotle for feedback and review.

I talked to 3 friends over the holidays about a model I have in my head for classifying all the media I consume because I’m that kind of nerd, and I want to share it because I think it’s useful for folks who care about content, whether that’s as a creator or as a consumer.

>It’s also just fun to make up graphics and pretend like I know things about stuff, so buckle up for some chart art.

The model is pretty simple. It combines two well-known spectrums together: the purist-tourist spectrum, which Virgil Abloh came up with and has been discussed in creative circles forever, and the timely-timeless spectrum, which is a marketing 101 idea. I’ll elaborate on both below, but here’s a visual to start.

Let’s unpack this a bit. On the x-axis, from left to right, we have the purist-tourist spectrum, which covers the range of target audiences that a piece of content might have. A purist is an expert or someone with domain knowledge. They’re high context people who can appreciate the finer things in their niche because they know what to look for. A tourist on the other hand is a low context person that’s new to a niche but keen to learn.

Side note: In this framing, “tourist” does NOT carry the negative connotation of not-belonging that it usually does on social media… instead, it’s meant to signal fresh eyes and curiosity.

Next up, on the y-axis, from top to bottom, we have the timely-timeless spectrum, which covers the range of how time sensitive a piece of content is. For clarity, let’s say that “timely” means responsive to the present discourse and “timeless” means evergreen.

When you slam these two spectrums together it creates 4 quadrants with different characteristics, and I labeled those with different grey boxes to denote the kind of creator that makes content at each intersection. I’ll define each of them quickly.

Clockwise from top left we have:

Insiders. These creators make content that’s timely for purists. They’re very plugged into the web of people and information in their niche, and they have access and insight that’s highly esteemed by other insiders.

Scouts. These creators make content that’s timely for tourists. They go deep into new topics like insiders do, but they’re focused on making complex ideas accessible to a wider audience.

Guides. These creators make content that’s timeless for tourists. They focus on curating the core ideas in a niche and explaining them to new people.

Artisans. These creators make content that’s timeless for purists. They go deep on topics and add original insight to their domain. Often this sort of content becomes canon and is referenced by other purists.

Let’s rip through 2 examples in real life to illustrate this further: tech media first, then crypto media second. Feel free to read both or skip to the topic you resonate with most.

Here’s a tech media mini map with some example brands/creators. Starting from the top left and going clockwise we have…

TBPN. The Technology Business Programming Network is insider media. Jordi and John host a 3-hour daily show covering The Current Thing in tech. They have a massive fan base because they’re high trust people that are widely connected and their content is very aware of the niche they operate in. It’s timely content for purists.

WSJ Tech News Briefing. This particular Wall Street Journal podcast is an example of scout media. They have good, bite-sized coverage of tech industry news that I can send to my college friends without context to help them understand current events. It’s timely content for tourists.

Wired: 5 Levels Series. My favorite example of guide media is this series that Wired does where they find an expert in a deep tech category like machine learning or zero knowledge cryptography, then they get them to explain their subject at different levels of complexity with an emphasis on education. It’s timeless content for tourists.

Acquired. This podcast is the epitome of artisan media. Ben and David publish ~10 episodes per year that spend 3+ hours each covering a well-known business like Google, LVMH, or the NFL. Every episode has a long shelf life and millions of people make time to listen whenever an episode drops. It’s timeless media for purists.

Next up, here’s a crypto media mini map with some example brands/creators. Starting from the top left and going clockwise we have…

CounterParty TV. Threadguy is one of the biggest steamers in crypto, and CounterParty is his company and daily show. It’s an example of insider media because he gets the best guests in the industry to come talk about current topics, and people tune in because they get to hear from folks who don’t otherwise do interviews. It’s timely content for purists.

CoinDesk Daily. As one of the oldest publications in the industry, CoinDesk publishes a lot of content, but their Daily series is an example of scout media. These videos cover what’s going in crypto in terms of regulation, price action, and company announcements in a way that’s accessible for people who just want to follow along with headlines. It’s timely content for tourists.

Coinbase’s Earn series. This video series was an entry point into crypto for a lot of people. It’s an example of guide media because back in the day it helped newcomers learn about the industry with the added incentive of letting them “earn” some coins in the process. Even though it was technically corporate media, I think it fits well as an example of timeless content for tourists. The videos are still helpful.

Cobie’s essays. Cobie’s essays are the best example of artisan media that I can think of in crypto. I’m not sure whether he thinks about his writing like that, but in terms of essays that get referenced as canon in the industry his stuff is at the top of the list. There’s something shiny about distilling a decade of experience into a few thought pieces. It’s timeless media for purists.

Now, there are lots of ways to analyze both the tech and crypto charts. We could geek out in 34 directions if this was an in-person chat, but to keep things tight, I’ll only share one pattern I notice before moving on, which is that purist media (in any domain) tends to be a point of reference where groups of people gather around the same podcast, essay, or video and use it as shared context… then on the flip side, tourist media tends to be an individual exercise in downloading knowledge like that scene in the Matrix where Neo learns martial arts by being plugged into a computer.

TLDR: Purist content is social in nature while tourist content is not.

Ok this is the part before wrapping up where I have to admit that there are some problems with the mini map we’ve been using. An obvious one is it’s rare for creators and brands to fit cleanly into a single category, and the grid struggles to capture that nuance. There are other good critiques as well – we’ll see if they come up in the comments – but friction aside, I like this model because it helps me make sense of media in different niches.

There’s a quote from an old statistician that’s relevant here which is, “all models are wrong, some are useful,” and that’s the thought I want to end on. Even though these things have pros and cons, I think forcing a model onto the world is a worthwhile exercise because that process helps you understand where it does reflect reality and where it grates up against it.

You learn just as much in either case.

-BR

P.S. Here’s an empty version of the mini map if you want to play around.

Thank you for reading my writing. I really appreciate your time and attention. If you’d like to get future essays in your inbox you can subscribe (for free) below.

Subscribe now

I grew up on the internet, and I love it, but something broke in me during the COVID era and its aftermath. I worked remotely in the crypto industry for 5 years, spent all my spare time on Twitter, and got stuck in a doom loop of e-addiction that I couldn’t unplug from. It fried my brain. So, I took a break this year to reset and reckon with my relationship to the world wide web (and to tech in general), and this post is a set of notes to share how I’m currently thinking about the internet and how to stay mostly sane while juggling the pros and cons of being online.

1. Let’s start with the obvious, which is that humans are more online than ever before. I could point to statistics, but I think this is intuitive.

2. The main distinction I see in my life from before and after COVID is that at some point the internet stopped being a tool I used every day and instead my online-ness became a default state of being. I was always on.

3. It’s become obvious to me that we’re not built to live terminally online lives. We’re not built to fill every gap in our day by doomscrolling, checking messages, and generally consuming global information streams at the speed of light. It’s sensory overload.

4. Being so online has implications for our personal lives. Common side effects include increased anxiety, depression, envy, and insecurity. That, along with shrinking attention spans and disrupted sleep.

5. Being so online has implications for wider culture as well. People are becoming more average over time because we’re all consuming the same content curated for us by TikTok, YouTube, Instagram, X and so on. This is often talked about as “algorithmic flattening,” and what that means is when we all consume the same inputs, we create similar outputs whether that’s in the form of art, startups, research, or even thoughts. We’re all losing our floofy spiky uniqueness and being eroded into a standard shape, which crushes originality and creativity at a societal scale.

6. Anyone who is very online understands these issues in a visceral way. We see it in our lives, and we know that being hyper online has some problematic secondary effects... and yet, so many of us have given up trying to set boundaries. We make excuses. I make excuses. “Oh, it’s just an hour and I deserve this time to scroll” or “Oh, the whole world is moving more online anyway so why bother trying to swim against the current?”

7. The reality is most of us are addicts. I understand this is uncomfortable language, but so many people have been completely defeated by the dopamine engineering of casino capitalism. Together, we’re willingly dripping poison into our lives instead of reckoning with our individual choices and the systems that we live in. We are, as Neil Postman said back in the 1980s, amusing ourselves to death.

8. This is not meant to be a drive-by guilting. And yes, in part, I’m talking to myself because I’m easily among the worst offenders in this category: I’ve done everything wrong and I lived in a dystopia of being plugged into endless, multi-screen slop at 2x speed for years. The reason why I’m writing this is I realized that I need to wake up and get my head in the game before my life is vaporized into a puddle of infinite digital regret, and I can’t imagine that I’m alone. I keep asking myself, “Don’t I want more than this? Isn’t life supposed to be more than this?”

9. Ok that was the worst of this crash out. Deep breath. All is not lost.

10. I concede that the internet is fun, funny, and a great place to learn. I concede that the internet is a magical tool for connection. I’ve made many of my closest friends online. I concede that the internet is an abundant place to find opportunity. Every job I’ve ever had as an adult is a result of having an online presence.

11. I think the supermajority of people would agree that there are a lot of great things about the internet, so what am I saying here?

12. I’m not saying that people should log off forever. In modern times that’s not even possible, and plus I like being online. I would never want to disappear completely. The question I’m wrestling with is, what’s the line? How much of the internet do we use and when do we say, that’s enough?

13. I’m not in the business of prescribing particular answers. A mentor once told me that most areas of life are morally neutral in the general sense, but that doesn’t mean they’re morally neutral for YOU specifically. For example, maybe one person can drink alcohol no problem, but another person can’t because they have an addictive personality. So, for internet stuff the question of what to use and what not to use (and in what measure, and when) is personal. We have to make those decisions ourselves.

14. That said, I think the through line is everyone would benefit from being more intentional with their relationship to the online world.

15. Some different practices that I’ve found helpful are limiting my attention to one social network, reducing complexity on my phone, and only doom scrolling and checking messages a few times per day. That, and a digital sabbath: one full day of no technology per week. Also, I do as much work on paper now as I can.

16. If you want to think more deeply about what good digital hygiene could look like for you then check out Cal Newport’s writing, Jenny Odell’s book, and the Center for Humane Technology. If you’re more philosophically inclined, then Sherry Turkle and Wendell Berry are worth reading as well.

17. For what it’s worth, I’m hopeful. I feel confident that I can claim my life back from being perma-online while still enjoying my favorite corners of the internet. I also think culture is shifting and pushing back against the default online-is-always-good position, which is encouraging.

18. One thing I am haunted by though is how much internet addiction is a system problem vs. an individual responsibility problem.

19. Let me give you an anecdote from another industry to help make this point. In the early 2000s, British Petroleum popularized the phrase “carbon footprint” as a way to individualize responsibility for carbon emissions and shift attention away from the largest corporate actors. It was a very effective propaganda campaign, and today when it comes to climate issues many people think about their personal carbon footprint vs. regulating the world’s largest emitters.

20. The question this brings up for wicked problems like climate change, internet addiction, or otherwise is… to what extent are people responsible for solving issues on their own vs. to what extent are large actors responsible for making changes (usually via regulation)?

21. The only answer I have is: probably some of both? I don’t mean to make this post too political, but the more I think about the modern internet and its impact on society, the more I wonder what regulating aspects of it might look like.

22. I know people panic when reading commentary about regulation because how dare you take away muh freedom, but I think Isaiah Berlin’s distinction of “freedom from” vs. “freedom to” is relevant here. Shouldn’t people, if they so choose, also be free from algorithms that have been optimized by the smartest people in the world to suck them in and siphon their attention?

23. At a minimum, I think everyone can agree that Big Tech is big business. We know the internet is not a neutral place and that algorithms are not neutral. This has implications across business, politics, education and more but those conversations are beyond the scope of this post (and mostly above my pay grade).

24. Alright back to some positive thoughts.

25. I do think individual choice is very powerful.

26. I would encourage everyone that read this far to explore what having a healthier relationship with the internet could look like. I know doing work on this has been worthwhile for me.

27. When trying to make changes in my life something I’ve noticed is that carrots are stronger than sticks, and by that, I mean spending time on activities that I’m excited about seems to be a better counterweight to overusing the internet than trying to force arbitrary limits on myself. I think humans are more driven by incentives than deterrents, so maybe the answer for all of us is simply that we need to fall in love with real life again. That’s what I want.

28. I realize a lot of what I’ve talked about can feel overwhelming, and it’s difficult sometimes to know where to start, so what I’ll say as I wrap up is this doesn’t have to be that deep. Stuff as simple as not having your phone in your room overnight or making sure to eat breakfast before logging on can be incredibly helpful. Small changes can have meaningful impacts.

I like to collect quotes. I have a decade worth of them sorted into files from different books, articles, movies, music, and other random places. It's a fun way to keep track of what I found cool or what I was “learning” in different periods of my life.

Over the last month, I went through all of these quotes, and I had this idea that the best of them together actually capture a lot of how I see the world. So, I spent some time curating this turbo gigantic list down to just the best stuff, and the result is this post that I want to share with you, which acts as a sort of personal operating system for my life (or a system-level prompt for how I think if I was an AI).

There are 7 sections: inner life, modern culture, conformity and agency, the social world, suffering, time, and mystery. It’s eclectic. It’s not really about technology or what I normally write about, and yet it feels like the most important thing I could publish this summer.

I’ve added some short commentary after each section, but I mostly want to let these quotes stand on their own. Feel free to skim and skip around. I hope you find something here that resonates or makes you think.

Inner Life

“The mind is its own place, and in itself can make a heaven of hell, a hell of heaven.” (John Milton)

“Fear is an illness. If you catch it and you leave it untreated it can consume you.” (Geralt, The Witcher)

“For myself, I am an optimist — it does not seem to be much use being anything else.” (Winston Churchill)

“Life is fucking electric bro. Don’t fall for the doomer shit. That’s for losers and normies scared of their own shadows. Walk around like God sent you and smile at everyone you see. Spread light and abundance. Build things and take chances. This is the best time in history!” (Goldie)

"There's a guy in my head, and all he wants to do is lay in bed all day long, smoke pot, and watch old movies and cartoons. My life is a series of strategems, to avoid, and outwit that guy." (Anthony Bourdain)

“I will have to remember, ‘I am here today to cross the swamp, not to fight all the alligators.’” (Rosamund and Benjamin Zander)

“Between stimulus and response there’s a space” (Viktor Frankl)

“I am not what happened to me, I am what I choose to be.” (Carl Jung)

This first section is about the mind and how even though our brains can lead us into chaos sometimes we can be intentional in the face of all inner and outer struggle. Our brain plasticity never stops changing, and this ability to act and choose how we respond is incredibly powerful. No matter how bad someone’s training data is, they always have the ability to switch out that data and tell themselves better stories through how they interpret and assign meaning to different events. Joy is a choice, and as that old Goldie tweet says, life is electric.

Modern Culture

“Our whole culture is based on the lie that it’s possible to be certain about the future.” (Barry Michels)

“The pursuit of data has become a form of substance abuse.” (Edwin Friedman)

“The trouble with market research is people don’t think what they feel, they don’t say what they think, and they don’t do what they say.” (David Ogilvy)

“All models are wrong. Some are useful.” (George Box)

“The map is not the territory.” (Alfred Korzybski)

“The real problem of humanity is the following: We have paleolithic emotions, medieval institutions and godlike technology.” (E.O. Wilson)

“The test of a first-rate intelligence is the ability to hold two opposing ideas in mind at the same time and still retain the ability to function.” (F. Scott Fitzgerald)

“When the facts change, I change my mind. What do you do, sir?” (John Maynard Keynes)

A lot of modern society is structured around the idea that humans are “rational” actors. This assumption is false for many reasons, and that’s why behavioural economists have been so relevant for the past two decades i.e. they challenge the rigidity of life being this perfectly logical system and introduce us instead to the messier and qualitative aspects of the human condition. The best way to engage in the world knowing that we’re operating faulty machinery (i.e our brains) is to have some humility, be open to new ideas, and be attentive to data but not controlled by it.

Conformity and Agency

“Most people are other people. Their thoughts are someone else's opinions, their lives a mimicry, their passions a quotation.” (Oscar Wilde)

“Rules direct us to average behaviours.” (Rick Rubin)

“I hate quotes. Tell me only what you know.” (Ralph Waldo Emerson)

“Everything around you that you call life was made up by people that were no smarter than you and you can change it, you can influence it, you can build your own things that other people can use.” (Steve Jobs)

“The world is a very malleable place. If you know what you want, and you go for it with maximum energy and drive and passion, the world will often reconfigure itself around you much more quickly and easily than you would think.” (Marc Andreesen)

“It had long since come to my attention that people of accomplishment rarely sat back and let things happen to them. They went out and happened to things.” (Leonardo da Vinci)

“Never doubt that a small group of thoughtful, committed citizens can change the world; indeed, it's the only thing that ever has.” (Margaret Mead)

“I can no longer obey; I have tasted command, and I cannot give it up.” (Napoleon Bonaparte)

So many people are stuck on the escalator of an average life across their career and their creative pursuits. This is heartbreaking because individuals and groups have radical freedom and agency and power to create things and make change in the world, but we so often miss this because most systems that we’re a part of teach us to conform instead. It’s intoxicating to move against this grain, to think for yourself, to tinker, to have a view, and to do your own things. To do that though, it requires moving away from convention and having the subtle defiance to chase what interests you regardless of other people’s opinions.

The Social World

“Everyday before my kids leave for school, I get them at the door and I’m like Lola, Zahra check this out: As soon as you leave this door, nobody gives a fuck about you. Nobody in the whole world gives a fuck about you. Nobody thinks you’re cute, nobody thinks you’re smart, nobody gives a fuck about your opinion. Nobody on the whole earth outside of this door gives a fuck about you. Nobody. And even some of the people inside the house... a little on the fence.” (Chris Rock)

"I'm at that stage in life where I stay out of discussions. Even if you say 1+1 = 5, you're right, have fun." (Keanu Reeves)

“No man is an island.” (John Donne)

“The story of any one of us is the story of us all.” (Fredrick Buechner)

“We must learn to live together as brothers or perish together as fools.” (Martin Luther King)

“Laughter is the shortest distance between two people.” (Victor Borge)

“We were together. I forget the rest.” (Walt Whitman)

"Treat people as if they were what they ought to be, and you help them to become what they are capable of being.” (Johann Wolfgang von Goethe)

Most people don’t really care about you or spend time thinking about you, and while that might seem unsettling initially, there’s a lot of freedom in accepting it and opting to live life on your own terms. That said, individualism has its limits. We need one another. There’s a lot of beauty in interdependence and community and laughter and friendship. Being intentional here makes us better, and it makes everyone around us better too.

Suffering

“I guarantee you that at some point everything is going to go south on you. You're going to say this is it, this is how I end. Now you can either accept that, or you can get to work. That’s all it is. You just begin” (Mark Watney, The Martian)

“He who has a why to live can bear almost any how.” (Friedrich Nietzsche)

"The most beautiful people are those who have known defeat, known suffering, known struggle, known loss, and have found their way out of the depths." (Elisabeth Kubler Ross)

“People fall so in love with their pain, they can’t leave it behind. The same as the stories they tell. We trap ourselves.” (Chuck Palahniuk)

“We cannot simply sit and stare at our wounds forever.” (Haruki Muakami)

"You never know what worse luck your bad luck has saved you from." (Cormac McCarthy)

“As I walked out the door toward the gate that would lead to my freedom, I knew if I didn't leave my bitterness and hatred behind, I'd still be in prison.” (Nelson Mandela)

“No prize fighter can go with high spirits into the strife if he has never been beaten black and blue; the only contestant who can confidently enter the lists is the man who has seen his own blood, who has felt his teeth rattled beneath his opponent’s fist, who has been tripped and felt the full force of his adversary’s charge, who has been downed in body but not in spirit, one who, as often as he falls, rises again with greater defiance than ever.” (Seneca)

Life is hard, and the strugglebus comes for all of us, but even in these moments people are surprisingly resilient, especially if they’re able to find meaning in the chaos. There is also incredible growth downstream of all forms of suffering… but many people get lost inside and never recover from their Big Bad Thing, so they don’t experience that evolution. The shocking truth is that no matter how horrific the things are that we go through, we always have the power to move along and let go, even if there are scars that stick around.

Time

"Fast gets all our attention, slow has all the power." (Stewart Brand)

“All we have to do is decide what to do with the time given to us.” (Gandalf)

“There are always plenty of rivals to our work. We are always falling in love or quarrelling, looking for jobs or fearing to lose them, getting ill and recovering, following public affairs. If we let ourselves, we shall always be waiting for some distraction or other to end before we can really get down to our work. The only people who achieve much are those who want knowledge so badly that they seek it while the conditions are still unfavourable. Favourable conditions never come.” (C.S. Lewis)

“How we spend our days is, of course, how we spend our lives.” (Annie Dillard)

“The content of your character is your choice. Day by day, what you choose, what you think and what you do is who you become.” (Heraclitus)

“A man on a thousand-mile walk has to forget his goal and say to himself every morning, 'Today I'm going to cover twenty-five miles and then rest up and sleep.” (Leo Tolstoy)

“You waste years by not being able to waste hours.” (Amos Tversky)

“There are decades where nothing happens, and there are weeks where decades happen." (Vladimir Lenin)

The pace of the world is relentless. We’re constantly fighting against information overload and algorithms that want to capture our time, attention, and money. In that environment, it’s important to be intentional: to realize life is short and precious, to choose what to spend that time on, and to commit. There is so much power in showing up, trying, and having basic routines. Most good work happens in small actions done every day that incrementally compound into bigger things, though there is also value in ignoring incrementalism to experiment and/or make use of inflection points where a lot of progress can happen in short periods of time.

Mystery

“Men are haunted by the vastness of eternity.” (Odysseus)

“Most experiences are unsayable, they happen in a space that no word has entered.” (Rainer Maria Rilke)

“The problem with pragmatism, as they say, is that it doesn’t work.” (David Brooks)

“Medicine, law, business, engineering, these are all noble pursuits, and necessary to sustain life. But poetry, beauty, romance, love, these are what we stay alive for.” (John Keating, Dead Poet’s Society)

“In the day-to-day trenches of adult life, there is actually no such thing as atheism. There is no such thing as not worshipping. Everybody worships.” (David Foster Wallace)

“God invented man because he loves stories.” (Hasidic saying)

“The most important thing to keep remembering is that life is not so much to be understood as it is to be lived out; it is not a problem to be solved, but a mystery to be participated in.” (Tim Hansel)

“Sacred or not it’s all that we got” (Shad)

There is so much beauty and so much mystery in being human. I should almost stop there and let these quotes stand on their own, but what I’ll say is I think we all have some transcendent impulse in us, some divine spark, some pull to direct our lives to an ultimate thing outside ourselves. I’m against nihilism. I find most postmodern thinking both intellectually unsatisfying and experientially false. Instead, I think there’s deep coherence and meaning available to all of us, and while so many of the great eternal questions are harder to pin down with more clarity than that, I tend to think there’s a god somewhere in the midst of it all. Though even if there isn’t… there is a lot of magic to be found together on the road.

I’ll stop there for now. 56 quotes is a lot. The hope in writing and sharing this was to open-source some of how my brain works, and if you read all the way down to the bottom here I hope it was worthwhile. There are other categories of life that I wasn’t able to hit so I might need to do a round 2 at some point in the future, but until then if you have other cool quotes to share or just want to riff on an idea that resonated, feel free to reach out.

-BR

Written collaboratively with Cheryl Douglass

Everyone on the internet knows that being active on social media is valuable, but a lot of people struggle to play the game. The real challenge isn’t about posting consistently – anyone can throw random thoughts out there – it’s more about how to find a genuine voice that offers a fresh point of view (and doesn’t come across as cringe or forced). This can feel like a hard problem for people to solve, but the answer basically comes down to one word: authenticity.

Now, authenticity can feel like a blah word/concept because it’s overused, so let’s get a bit more specific here. Authenticity doesn’t mean sharing everything in your life, unfiltered. What it does mean is thinking for yourself, having a clear voice and point of view, and being consistent in how you engage with others over time. Often the best and most widely loved accounts are just people who share their thoughts publicly and honestly about things that matter to them.

What we'd like do with this essay is talk to people who want to be more active in this way on any social network, whether you’re starting from scratch or trying to resurrect an old account. The game plan is to riff on some concrete suggestions that we’ve found helpful, both for ourselves and in working with founders at Seed Club. We’ll also cover two quick case studies.

Start where you are. The best way to start being active on the internet is to talk about things that you’re already doing or thinking about. If you’re a founder, talk about the founder's journey. If you’re an artist, talk about the creative process. If you’re an investor, talk about interesting opportunities in the market. And so on. Share process, not just outcomes.

Experiment. It takes time to figure out your voice, sense of humor, best content formats etc. If you just show up and try, that puts you in a small group of people who are actually active, and there’s so much to learn in doing that. A lot of getting comfortable on social media just comes down to reps.

Embrace your quirks. You’re special. Believe it. The best accounts out there double down on the weird things that make them who they are, and they naturally filter their ideas through their unique way of being in the world. Quirks are a superpower.

Define your core principles. What do you stand for? What are your values? What do you want people to think of when they see your username and pfp on the timeline? Answers to these questions help inform how, where, and when you engage.

Focus on a niche. Interact with people who are doing things that you think are cool. Make lists, be specific, and focus on a couple particular areas to start. Whether it’s fashion or derivatives trading or whatever else… it’s better to start by being deeply knowledgeable about a couple of categories than superficially trying to engage with everything out there.

Adapt to platform context. Instagram, TikTok, X, LinkedIn, Farcaster, Lens etc. all have different norms. How you engage in each domain should take that into account. “Local” knowledge and vibes matter, and your voice and content should change a bit in each place accordingly.

Be human. Engage with people. Post stuff but be willing to be wrong publicly. Pay attention to the conversations and themes that are getting talked about, then engage. It’s much more natural to participate in The Current Thing than just randomly posting.

To make all of this more real, let’s look at two case studies for some inspiration. One on the personal account side and another on the brand account side.

Personal Case Study: Winny

Winny has crafted a cool personal brand across multiple platforms. Her content covers fashion, tech, crypto, travel, lifestyle brands and other themes. Her tone is casual, irreverent, and playful – it’s like having a conversation with a friend. She founded a company called Chipped which sells NFC-embedded nails as both a beauty accessory and a tool for digital and social interactions. She has helped grow the brand massively through her own content.

What sets Winny's approach apart is her commitment to genuine vulnerability. She isn't afraid to show the struggles that come from entrepreneurship alongside the wins, and this creates a relatable narrative that resonates with her audience. Her content strategy emphasizes personal expression and lifestyle, speaking directly to an audience of millennials and zoomers that value authenticity. She uses a mix of product showcases, behind-the-scenes content, founder journey moments, user generated content, celebrity collaborations and more, all filtered through her personal voice, which makes people connect and align with Chipped in a way that makes it feel more like a movement than a product.

Brand Case Study: Duolingo

Brand accounts are a slightly different conversation than personal accounts, but Duolingo is a fun and instructive example to call out regardless. That team really cracked the code.

In short, the Duolingo social strategy combines deliberate chaos with platform fluency while having a consistent brand everywhere their owl character goes. Duo’s unhinged personality is carefully calibrated absurdity that resonates with Gen Z humor while staying true to the company’s core mission of making it fun to learn languages. And the team doesn’t just post content, they really speak the native language of each platform: TikTok gets chaotic skits, LinkedIn gets growth metrics, X gets snappy comebacks etc. This enthusiastic, slightly threatening, simp of an owl has created a lot of attention and brand value for the company across a wide variety of content.

In closing here, whether you're an individual creator like Winny or a brand like Duolingo, the most compelling online presences have a liveness to them. They feel real. In an era of infinite content, we get hit with endless slop… so to stand out you have to press into the perspectives and quirks that make you unique. The most magnetic accounts don't try to be everything to everyone, they choose depth over breadth and originality over optimizing for algorithms. It’s a game of authenticity + consistency + time.

No one really reads anymore.

In part, I think this is because we’re all flooded with memes, messages, and internet chaos every day, so our attention spans have been whittled down to nothing. In part, I think this is because there is so much quality content out there to consume. Like, why would you read when you could spend your free time scrolling social media, binging TV shows with decamillion dollar budgets, or shuffling through all music ever made?

Now, I don’t think it’s totally over for writing. At least I hope it’s not. But one thing is clear to me: if you want real humans to read what you have to say in the internet era, it has to be ferociously high quality. You need to figure out how to hook people who have broken (and waning) attention spans, then keep their attention while competing against the entire entertainment-industrial complex for a slice of their time. Hard to do, but not impossible.

I think three things matter if you want to publish work at this level i.e. writing that gets people reading… not skimming, not reading the opening 5 sentences and closing the tab, reading, whether it’s a tweet, an email, a thought piece, whatever. It comes down to understanding your audience, having a coherent structure, and writing with your own style. Let’s riff on each.

Audience. All good writers have an opinionated view of who they’re writing for.

This could be high level, like writing for all basketball fans. It could also be hyper specific, like writing for people who are obsessed with Tubby Cat NFTs. The point is: having something to say when you write isn’t enough. You have to adapt content to a particular audience because those constraints are what make the writing come alive.

Let’s take Anthony Bourdain’s memoir Kitchen Confidential as an example. The book hits so hard because he wrote it for the restaurant industry, and he did it with an insider’s tone, quirks and sense of humor. What made it so high quality was it was so high context.

The main writing lesson from that book is to be clear who your readers are. If you can write about whatever your topic is with relevant references, experiences, and terminology then other similar-minded folks will be drawn in. Ultimately, attuning what you say to a specific audience is relevant for everything from low register shitposting on Twitter to high register research for academic journals.

One way I try to integrate this thinking into my essays is by obsessing over accessibility. I know I’m writing for busy people in tech, finance, and creative industries who (maybe) trust me with 5-7 minutes of their time while procrastinating at work, so everything has to be casual otherwise most of you won’t bother reading.

Structure. All the best writing presents ideas in an order that just makes sense.

In his book Draft No. 4, John McPhee says, “a piece of writing has to start somewhere, go somewhere, and sit down when it gets there.” What I take from that quote is, in all cases, good writing is a game of creating good structure. The challenging thing about structure in writing though is it’s not a math equation. There is no single correct way to write anything, and writers need to figure out what to say, when, and why.

In wrestling with those trade-offs, there is one “rule” that I come back to (also from McPhee) which is letting your ideas about a topic create a structure for you instead of trying to force something from the top down. For example, I’ll collect ideas about a topic, compile them into digital bullet points, and pull those ideas around until something makes sense. This story works here, this anecdote works below that, this joke needs to get cut… it’s an emergent process. Eventually, you hack out a rough outline, and voilà.

One caveat: people can for sure shoot from the hip and just write. Often the best work comes out in fits of emotion, inspiration, or good vibes. But even then, you can usually triage your rambling with some restructuring and make it more coherent.

Style. All elite writers have a unique voice. It’s what separates the great from the good.

Take a second and think of a book or essay that stands out in your memory. If I had to bet, it probably got stuck there because of the author’s style. It’s the way David Foster Wallace talks about tennis and the way Joan Didion talks about illness that makes their work so electric to read, less the subject matter itself.

I think cultivating your own writing style is a lifetime journey for everyone who likes to move words around on a screen for fun, and it’s impossible to really tell other people how they should create something that is so personal, but beyond just reading and writing a lot and getting the reps in there are two things I’ve personally found helpful here.

First, focus on being authentic. Consume content you like, then filter ideas about those topics through your own voice. Whatever you can do to inject more personality into your writing will help it stand out. Second, have fun on the page. No one wants to read another eye-bleedingly boring ChatGPT output, and people can tell when writers are jazzed about their own work. It pulls you in.

To close out here, I want to say that I’m confident writing has its place in the digital age. People will always be readers because writing is just thinking on paper (or screens), and good thinking always has an audience.

That said, it’s already tricky to compete for attention in the social media era, and over time the torrent of content we all consume daily is just going to get more insane, so writing has to get a lot right to cut through the noise.

The best way I know how to execute at that level is by obsessing over who you’re writing for, what order your ideas are presented in, and how your content is filtered through your own voice and life experience. Audience, structure, style. A.S.S. I think if you get those three parts right, everything else falls into place.

I think anyone alive today would agree that attention is an insanely valuable resource. There are a finite number of hours in a day, and how all 8.1 billion human beings spend that time makes the world what it is, from consumer spending to politics.

In business, we see this idea packaged as the “attention economy,” where the core objective of so many companies is to capture a consistent chunk of our time. Everything from Spotify to Netflix to Fortnite to Paris Fashion Week to PornHub to the Olympics are all ultimately playing the same game, and it’s a cross-domain war for who can collect, monetize, and harvest the most human attention.

In financial markets, there are similar dynamics at play where flows of value are driven as much by memes and stories as “fundamental” metrics. Think about the COVID-era run on GameStop shares, or Tesla as a meme stock being worth more than the next 7 largest car companies combined, or the 2021 crypto bubble with all its dog coins, and so on.

Attention makes the world go around.

Now, this has always been the case to some extent, especially at the individual level. People who can command attention have always had disproportionate influence on the periods they lived in, including everyone from Cleopatra to Martin Luther blah blah, BUT there is something supercharged about the post-2010s social media era where all of us are a few clicks away from billions of people.

This is probably most intuitive at the celebrity level where, say, Mr. Beast as a YouTube creator or AOC as a congresswoman have insane reach compared to legacy actors and politicians.

What is more interesting to me than the elite part of this conversation though is how the same shifts in technology that have empowered celebrities with such huge reach have also created the conditions for an attention “middle class” to emerge i.e. people who can command some amount of attention within a niche but aren’t A-list celebrities.

I’ve started to think of these folks as attention millionaires.

An attention millionaire is someone who can command 1 million or more minutes of human attention over the course of a calendar year. This is the equivalent of 16,667 hours or 695 days’ worth of attention that someone has to capture every year to be part of the club, whether that’s through publishing a newsletter, dropping short-form video content, tweeting, or whatever else. Stay with me.

First, why a million minutes? I don’t know I made it up lol, but it is a helpful concept because it gives us a way to classify a different sort of person between the average consumer that doomscrolls brainrot content for hours and the Kardashian-level super moguls of the world that dictate the direction of culture. In a way, what I’m talking about is like a “micro influencer” in scale, but that term isn’t quite specific enough.

Second, why does this group of people matter? In short, they matter because with that amount of reach a creator (of any kind) has hit enough critical mass to have an impact, both on the world and in their own lives. With a million minutes under your control, you have enough attention to get jobs, start businesses, help friends raise money for charity, contribute to zeitgeisty conversations, make supplemental income on referral fees for different products, etc.

Over time, I think people in this category will collectively have more influence on the world than celebrities do in the same way that economies are more driven by middle class income earners than high net worth folks.

Think about it. Who has more direct and tangible impact on your real life:

Gordon Ramsey? Or the dude that writes a laugh-out-loud food blog where you get your local restaurant recommendations from? Diane von Furstenberg? Or your friend’s cousin that runs a fashion TikTok and is an early adopter of every trend that matters? Warren Buffet? Or the high time frame speculator you follow on X (with notifications on) because he makes good calls about markets that you align with?

It’s not even a contest in my mind, the niche people win every time because they have a stickier connection to us and our daily lives and a stronger claim on our attention. And as society continues to orient itself more intensely around attention as the core primitive in business, politics, and culture the attention millionaire class is only going to grow in power and relevance.

To me, they’re the real main characters of the internet era.