Coinbase 的万物交易所：智能体金融、稳定币与资产代币化——对话 CEO Brian Armstrong

The minimum fee typically on a debit or credit card transaction is about a 30-cent flat fee, plus the additional percentage. Arguably, it doesn't really work well for anything under a dollar. Certainly, it doesn't make sense to pay 30 cents to send a 1-cent payment. I believe the stats we last saw are that about 76% of the agentic commerce transactions we're seeing are under 30 cents.

We don't want the AIs to be unbanked. We want to bank the AIs. They deserve financial services as well. The AI agents themselves are going to have to have their own accounts, and so we've made some simple tools where, with a single command, you just paste a prompt into your AI agent and it can now have its own financial account. It'll be interesting to see how the agentic economy evolves. There'll be more agents than humans in the not-too-distant future, and it stands to reason that the agentic economy will be bigger than the human economy at some point in the not-too-distant future as well.

1. Brian Armstrong Introduction

Today on No Priors, we are joined by Brian Armstrong, the co-founder and CEO of Coinbase, one of the world's most important and earliest crypto companies that has also been quite aggressive at adopting AI. Brian is also the co-founder of New Limit, which is a anti-aging and longevity company doing really interesting things in biology, including certain types of high throughput screening to understand uh aging, different uses of AI, etc. Brian is one of the most forwardthinking people I know. on different aspects of science, technology, art, the world, etc. So, it's very exciting to have this conversation with them and we'll be talking about everything from what's happening in the crypto world to what's happening in AI to what's happening in biology to how different individuals can have an impact over time. So, it's a very fun conversation.

Brian, thank you so much for joining us today. It's a pleasure to see you.

Yeah, good to see you a lot.

I was hoping we could talk about a few different topics. I think there's a bunch of stuff relevant to Coinbase and crypto, and agentic finance and things like that. But you also started a company called NewLimit, which is focused on longevity, aging, and biotechnology. I think it'd be great to cover both as part of this conversation.

Yeah.

Maybe we start with Coinbase. Coinbase has been making some really big moves. You've been floating this concept of the everything exchange as you've been building out what I think is a new global payments layer. Can you tell us, at a high level, what are some of the directions of the company and some of the areas you're most excited about right now?

We're at a really exciting time in Coinbase's history because, on the one hand, every asset class is coming on-chain. It's now possible to trade everything in one place with good liquidity and good cross-margin. For markets and trading, everything out there—stocks, commodities, crypto, derivatives, perpetual futures—it's all coming to one place, where traders can do everything in one spot, including prediction markets. They can do that in more countries now, given the regulatory clarity that's emerging, which we can talk about if you want.

The everything exchange is kind of our bread and butter, our core business. The next big area that's emerging is stablecoin payments, which have just been growing like crazy. Even when Bitcoin prices are down, stablecoin payments just keep growing. We're helping companies all over the world integrate stablecoin payments, which are fast, cheap, and global. You can send money at the speed of information—under 1 second and under 1 cent in the U.S. to move money anywhere in the world.

The third big area is something I know you talk a lot about on this podcast: how AI is changing everything. AI is changing everything, and so it's obviously going to change financial services, too. We call this agentic finance, or AIFi, and we're building this into Coinbase in a number of different ways.

The first way is that people who have an existing financial account are going to want to have an AI adviser. It's kind of like how wealthy people are able to get access to investment advisers, but everybody should have access to them now. They can help teach you about building the right portfolio, dollar-cost averaging, tax-loss harvesting, and all these kinds of things.

If the market dips 5%, 10%, or 15%, you can lean into this kind of portfolio over time, try to save money on fees, and earn yield while it's sitting in cash. You can just talk to it like a human, and it'll do all these things for you. At least that's the vision, and it's getting better and better. We've launched an adviser in the Coinbase product.

The next thing is that the AI agents themselves are going to have to have their own accounts. We're on a mission to go out there. We don't want the AIs to be unbanked. We want to bank the AIs; they deserve financial services as well.

We've made some simple tools where, with a single command, you just paste a prompt into your AI agent, and it can now have its own financial account. It's using a self-custodial wallet and crypto rails, so there's no KYC process. Our AI agents don't have a government ID. They can't walk into a bank branch—at least not yet, without their humanoid robot companion.

Now any AI agent can have its own financial account to make payments and pay for goods.

Is that tied to an account of somebody who's already on Coinbase? In other words, is this sort of like a subsidiary account, or is it an entirely new thing? How do you think about the hierarchy of that?

We're doing it both ways because I think there's going to be demand for both use cases. It's going to be possible to create a separate agentic account with segregated funds on Coinbase in the near future, and you can give your agent access just to that account to do trading and payments on your behalf. That's linked to your identity as a human.

We're also going to provide self-custodial wallets to any AI agent out there that wants to hold a balance in stablecoin. It could actually raise money or issue a token.

What's fascinating is that today we often talk to these agents, and it's sort of one person to one agent. They go get some work done, but they often get stuck. There's a paywall: “Can you put in your credit card? I need to spin up these AWS resources.” Or you wanted me to book this hotel for you, but here's the webpage—can you go put in your credit card? That kind of thing.

2. Agentic Commerce

In the future, I think you're going to want to have these agents have their own spending accounts, kind of like having a personal assistant or an employee who has their own spending account. They're probably going to need a stablecoin-backed credit card for backward compatibility, but we're seeing a lot of agentic payments now flow over these new protocols, like X42, which is one that we incubated at Coinbase and put into the Linux Foundation. Google, Cloudflare, AWS, and a bunch of other companies are working with us on that now.

That's been really cool. We put together a little fun website called agentic.market, where people can see the stats on how many payments are flowing through this new economy and see all the different services.

We're helping a lot of businesses become enabled in the agentic economy. Is your store ready to accept agents as customers? We have a product called Coinbase Business that helps companies accept stablecoin payments. That's good for humans who want to pay with stablecoin, but it's also good for agents, and they get both out of the box.

You mentioned things like stablecoin or crypto-backed credit cards. How do you think about what proportion of the future volume in the agentic future is going to be crypto versus more standardized financial rails?

What's interesting is that a lot of the agentic commerce that's happening involves pretty small transaction sizes, sometimes just a few cents. As you probably know, the minimum fee typically on a debit or credit card transaction is about a 30-cent flat fee, plus the additional percentage. Arguably, it doesn't really work well for anything under a dollar. Certainly, it doesn't make sense to pay 30 cents to send a 1-cent payment.

I believe the stats we last saw are that about 76% of the agentic commerce transactions we're seeing are under 30 cents.

What sort of transactions are those? What's happening at that low of a price point?

A lot of it is gathering information. You could think of it as venture firms doing research and compiling data from paywalled sources, or recruiters scraping data from LinkedIn. Oftentimes, it's actually agents talking to other agents that are specialized in some way and getting source data from them, almost like a tool call that's happening.

On the agentic.market website, we have a list of the top service providers that are doing things like search and financial data. I think what'll be interesting is that, in the future, you'll see other agents in the market that become specialized.

The frontier models are trained to be quite good generalists. They're good at a lot of things. We're seeing instances where smaller open-weight models can be fine-tuned, or can use reinforcement learning, to outperform the frontier models on a specialized task within a company.

I'm sure you've talked to lots of startups about this kind of thing, but it could be a compliance review queue. If you train a model on 100,000 compliance cases using Coinbase's proprietary data, a small open-weight model can actually outperform a frontier model.

And so I think you're going to see in this marketplace a bunch of specialist agents. It might be like, “Hey, this agent is just really good at ordering you pizza, if that's what you want.” Or this agent is an incredible designer, and that's all it's been trained on—tons of proprietary Figma data or something like that. So it'll be interesting to see how the agentic economy evolves.

It's probably common wisdom at this point among your listeners, but there'll be more agents than humans in the not-too-distant future. It stands to reason that the agentic economy will be bigger than the human economy at some point in the not-too-distant future as well.

Yeah, it's interesting. I think one of the safetiest arguments is the concern that there will be resource aggregation by AI, and then that'll be used to manipulate society or things like that. It's always interesting to think through some of the implications of the scale of these things and how they work. I don't share that concern as strongly. I'm just saying that's something that comes up from time to time.

Yeah. Well, just actually on that, it's funny. I've heard people like Elon say that they think there won't be money in the future, right? I actually agree with Elon on most things, but that one I was like, okay, I think I know what he means. He's basically saying at some point you start to measure things in energy and mass, and those become the scarce resources. I think, in the limit case, he has a point.

3. AI and Crypto

But the thing is, there's going to be scarcity for a long time. Even if the cost of goods and services really falls—if we have price deflation because robots and AI start to make commodities, food, and housing very cheap—that could happen. The government printing might also create a bunch of inflationary pressure. But there's always still going to be scarcity of certain things, like land, especially usable land in various ways. Energy, chips—Dyson spheres are going to be expensive, I think, pretty far into the future, so I think we're still going to need a medium of exchange.

It's really cool to see that, as AI emerges, crypto rails are just kind of a nice complement for all that stuff to finally work. Crypto was really good for humans, and it's going to be essential for AI.

4. AI Inside Coinbase

Yeah. People, I feel, have been talking about this for a while too, particularly in the early AI community. It overlaps so much with the crypto community, right? You'd see computer science majors coming out of MIT and, depending on the year that they graduated, they would go into crypto, and 2 years later they're all going into AI. It was roughly the same people—people who were really into certain math or algorithms or other things.

So there's an interesting bifurcation of people's careers based on when they graduated, in terms of whether they ended up doing crypto in 2020 or whatever it was versus AI in 2022.

Yeah. Well, I'm reminded of that meme. Have you seen that meme that says, “¿Por qué no los dos?”

You can do both. They're going to intersect here.

Yeah, it seems like they just might.

The other thing you folks have really been focused on is the use of AI within Coinbase itself. I know there's obviously coding and other things, but what do you think are some of the more interesting projects you've done internally?

Yeah. So, like a lot of companies, we started off with the basics. We got some coding tools. We got them hooked up into our various data repositories. We've been using it for a long time in machine learning for risk prevention and fraud prevention, and customer support inquiries started going through AI. So that was all good table-stakes stuff.

The things that we've started to do recently, where we're pushing the boundaries, are centered around recursive self-improvement. We basically want to get as close as we can to—and actually achieve—recursive self-improvement. What does that mean in various ways? It means that you have a brain for the company that can have a whole bunch of subsets, by the way. You can have a brain for each team, a brain for each service or repository within the company, and a brain for each individual, which, by the way, I believe they should be able to take with them when they leave Coinbase. So we're creating some good policies on that.

Let's say you have a service, like a GitHub repository or some backend service that's running, and today we have engineers spinning up agents to make changes to that. What we're just now launching and seeing recently is that every time the agent comes in to make a change to that service, it's actually ingesting the brain for the team working on it and for that repository.

And so what is that brain? It's basically a history of all the incidents that have ever happened with that service, all of the financial controls that have to be enforced with that, every A/B test that we've ever run on that, and, of course, the full GitHub history of pull requests that were accepted or rejected. The people running that team can go in there and look at it at any time. Today it's just Markdown files, right? It sounds kind of simple at the end of the day, but getting all of this hooked up is the key.

So let's say you ask the agent, “Hey, I want to make a change to this service.” It ingests the brain, makes that change, and then, as a human, you go to review it and you're like, “Ah, it missed something.” The simple thing someone might do in that moment is make a manual edit to fix it and then ship it.

But the key step that we're trying to really enforce now in our software factory, or software development lifecycle, is that when you make that change and edit the agent's thinking, that context has to go back into the brain so that you not only fix it in this case, but in all future cases going forward. Then you start to see the accept rate for one-shot PRs or that kind of thing just tick up over time, and you've now got a recursive self-improvement system.

That's one of the big ideas we've been working on recently, along with getting our own internal agent harness—which we call Toshi—integrated with all these tools. It can now make calls out to external vendors and pay using the agentic finance economy over these x42 protocols. If you want to edit the brain as you're making a change to that PR and have it go back into the brain, we've put that into the harness.

Asking it to do things like this is kind of amazing. Just earlier today, I was making a plan for this relatively complex feature. It's now even allowing people like me, the CEO, to get back in there and ship features.

But I asked one of the high-end, expensive models, “Okay, go make a plan for how this feature might be implemented. Break it down into 3 phases. Each phase has 10 different pieces that need to get built.” And I was like, “Okay, go execute phase 1. Go spin up 10 separate agents to go execute each of those pieces in parallel. And, by the way, recommend the right agent to use, because they might be cheaper models.”

So it spun up a bunch of open-source models and a bunch of Grok. Two minutes before I hopped on this podcast, it pinged me and was like, “By the way, all 10 of those things from phase 1 are now done and ready for you to review.”

5. Productivity and Company Size

Anyways, I'm starting to get to this place where it's kind of addictive, actually, because in the past, every time I'd have these ideas or bugs I'd find, I'd just be pinging the team all day long: “Hey, can we do this? Can we do this? Can we do this?” And now, instead of pinging the team in Slack, I can basically tag an agent in Slack with that comment or drop it into our internal harness.

Instead of bugging the team and asking them to do something, I just send them a PR for review, and it's already done. That's a magical moment that I find very addictive.

Yeah. I feel like the people who are best at adopting this, who aren't day-to-day engineers—and I know you had a long background working as an engineer at Airbnb and other places, obviously helping start Coinbase—but I found that many folks who do best at this, who are nontechnical, also approach it almost like a video game. You're leveling up in terms of your capability set and how you learn to do things, and the recursive loops you start building for yourself and everything else. So I think it's fascinating to watch that adoption.

How do you think that impacts how companies are set up in the future? In other words, some people make the argument that the average company size over time should get much smaller. It could be a startup, it could be existing companies, et cetera, simply because you're able to harness so much of this intelligence, so much of this proactivity, or this ability for people to do so much individually.

Do you think that's true? Do you think that's the direction things are heading? Is there a timeline where suddenly every company is 10 people or whatever, or do you think there will always be these larger-scale organizations?

Well, productivity is definitely increasing, right? So that's slightly different, though, than saying the average company size will be smaller.

6. The Everything Exchange and Tokenization

It's kind of like just saying tasks are being automated or eliminated. People are not being eliminated; I think tasks are being eliminated. If you take any sort of company today—Coinbase or others that exist today—it's not that I think we're going to shrink the headcount to 10 people or something like that. I think the existing people we have are going to be able to get much more done, and the pace of everything will accelerate. Now, it's also true you're going to see some companies that are 2 people or 5 people or whatever, and they're able to do a lot more than what we would have historically thought was possible. That's also true.

So it's increasing the velocity of the existing team—

And making things ship faster, with things automated. Basically, you're getting more scalability off of people. If you have a cost basis relative to revenue, then you suddenly have the ability to do dramatically more with that same group of people, which means, to some extent, your margins should start to rise over time if you're effectively growing revenue faster than you grow the team, although I guess some people may still grow them at the same pace.

Yeah. That's super interesting. And then, I wonder if it's worth spending a little bit of time as well on the Everything Exchange. I just think some of the things you're doing there are fascinating. There's prediction markets there. There's all these really interesting directions. What proportion of Coinbase, either volume or revenue, is sort of those other ancillary areas now outside of crypto?

Yeah. Well, we actually shared in our recent earnings deck—and those are the only numbers I'm probably allowed to talk about—that 88% of our revenue is from non-Bitcoin trading at this point. That's an interesting stat. I think it's funny because if you look at our stock price, it does trade fairly correlated with the price of Bitcoin.

I love Bitcoin. I think it's going to be around forever. It's digital gold. It's sound money for everybody in the world. I think it's amazing. But I also think all these other things are part of how crypto is updating every aspect of the financial system.

We just announced our tokenized stocks product, which is only available outside the U.S. right now, unfortunately, but we're working with the SEC and others to get a path to do it in the U.S. It's the first truly tokenized stock product. It's not a synthetic or derivative or some debt instrument; it's actually a security, and it's one-to-one represented by the actual security held in custody.

This is broadly a trend called tokenization. Stablecoins were the first major use case of this, with a token that represents one-to-one a dollar in a bank account or a U.S. Treasury. Now we're seeing it with stocks, which is really exciting. There are something like 4 billion people in the world who don't have access to any brokerage or U.S. investment account. They don't have any ability to invest in Coinbase, NVIDIA, Apple, or whatever.

Much like Tether was very successful outside the U.S. because a lot of people wanted dollar-denominated accounts, there are 4 billion people who want access to high-quality investments, and they don't have that today. So we'll do the tokenized stock piece. Then I think you just go on down the list, like private credit, treasuries, and bank deposits. Everything's going to get tokenized, and it's just going to mean that financial services is going to have modern rails to run on.

Of course, these can all be traded on the Everything Exchange. They can be sent back and forth. If you want to gift a share of stock to your nephew or something like that, you can just send it to their wallet. You don't have to go through some complex stock transfer process. It's antiquated. Money can now move at the speed of information, just like a WhatsApp message, and crypto is updating all those rails.

7. Prediction Markets

That's pretty amazing. It's basically broadening and modernizing the financial system, but in a very broad way, right? It includes, to your point, eventually bonds. It includes a subset of stocks. It includes all sorts of things that before had much more friction in terms of using them or purchasing them, and it opens it up for the world.

There were counter-versions of this in China and other places where, if something was listed on the Hong Kong stock exchange, sometimes these funds would almost be set up so that people could buy proportions of them to participate in some stocks that were listed in other countries. It'd be interesting to watch how global capital flows change as these things get bigger and bigger. To your point, it feels like the stablecoins in Tether were the first version of that with USD. It's kind of fascinating. I remember, actually, 7 or 8 years ago, I invested in a company called Harbor that was trying to tokenize real estate.

I think that's a little bit too far ahead.

Yeah, I think that'll happen too. I think you were early—early but too early—but right on that one. So it's kind of fascinating to watch those play out. How do you think about prediction markets? A lot of the use cases that have been mentioned in the past seem to have roughly consolidated around sports and a few other areas. Obviously, there are things in politics. Do you have a view on where, in the long run, prediction markets have the biggest footprint?

Well, I'm very excited about prediction markets. They're growing super fast inside the Coinbase app. I think we shared in our last earnings call that, within months of launching, it had hit a $100 million revenue run rate. It's growing really nicely, like over 100% quarter over quarter, something like that. I'd have to go look at the exact number again.

It's true that sports has been a part of it. I think that it's much bigger than that, though. That's great; that's a form of entertainment. We should celebrate that. I think it actually has much broader implications from a societal point of view that people maybe haven't fully appreciated.

Political elections are something that have already happened in prediction markets. But let's say that you want to try to understand what policies should be implemented from a government point of view. You could actually say, "All right, if we were to implement this policy, what would the unemployment rate be within 1 year, 2 years, 3 years?" Then a prediction market could form on that. It could help inform what policy is out there.

8. Introducing New Limit

You could also ask people the big questions in life: Is there a God? Were we designed by evolution or intelligent design? Some people would call these more like opinion markets. It's not that there's ever going to be a resolution date where we definitely know for sure on some of this stuff. If we did, it would be very far in the future and hard to imagine.

But if you ask somebody a difficult question, and there are events that are always unfolding in the world that might shift things—maybe 60% of the population believes X, but a new scientific paper comes out or a new terrorist attack happens, or who knows what, and then it shifts from 60% to 80%—people could still make bets on this.

It's kind of like when you invest in stocks. There's not some resolution date for the stock where it's done and you get your payout from NVIDIA. You're just betting on whether it's going to do better or worse in the future. I think you could imagine a whole other category, for instance, of prediction markets for the big life questions and things like that. We're just scratching the surface.

Another thing that you've been working on is NewLimit, which I think is really fascinating. When great founders do an act 2, and you were doing an act 1 while simultaneously doing an act 2, how did you think about starting a company while running a public company?

Yeah. Okay, so this is a really interesting topic because, like you, I'm interested in lots of different ideas. I think that's part of the benefit of being an investor, by the way: you get to talk to lots of smart people about a million things. So it's very intellectually stimulating.

Maybe the downside of it is that you don't ever feel like you're all in on one thing and you're in the arena sometimes. You might be sitting there from the sidelines saying, "I know exactly what they should be doing. Why don't they just do this?" If you're the CEO, you actually get to decide these things, for better or worse.

That's a long way of saying I like being a builder, or an entrepreneur. I'm probably better as an entrepreneur than I am as an investor. Some of my investments are a little bit more like, "Wow, that would be so cool if that happened." But once in a while, someone has to remind me, "You do realize the point of investing is to make money, Brian?" Okay, so I digress.

Coinbase went public in 2021. We were very profitable; it felt like a 10-year journey. I thought, "Do I want to do something different? Do I have another thing, or am I just at the beginning of this?" And I realized I'm just at the beginning of the journey with Coinbase. I think our mission of increasing economic freedom is just—we're just getting started.

About 0.5% of global GDP is running on crypto. We've got a lot of the job that's not finished, right? But I do want to have the ability to try some other ideas, maybe, and help incubate some companies or get them off the ground, whether I run them or not.

What I did was sit down at some point and ask myself, “What are the big technology trends that I’m excited about?” I think, by the way, if you make some money in software, you kind of have an obligation to put it into hard tech and help some of the bigger things that might advance society, which tend to be more capital-intensive. I was a little bit inspired by Elon doing these kinds of things with PayPal.

I sat around with some of my friends, and we asked, “What are some of the big trends?” It’s space, AI, fusion energy, brain-machine interfaces. There are good teams working on those. I didn’t have a strong thesis about what I would do that might be different.

When you’re coming up with an idea, you want it to be something that needs to happen in the world, but also something where you think you can uniquely add value. I didn’t see that many great teams working on longevity. I thought a lot of it was kind of snake oil.

You came from a bio background, so I cornered you and asked, “What’s going on with this space?” You were like, “All the science is a little bit bogus. People have tried these things for a long time, like caloric restriction or whatever. We found a bunch of ways to make mice live longer, and it doesn’t work in humans.”

I said, “All right, well, let’s host this dinner and see if we can invite some of our smartest CEOs and scientists.” You attended, I think, one of these dinners, and I really appreciate you inviting some people. We went around the room and asked all of them, “What’s the most exciting thing on the horizon from a bio point of view that you think is high-potential and underfunded today?”

People had all kinds of ideas, but one of them was this epigenetic reprogramming idea, which some labs had seen early results with, like Shinya Yamanaka’s lab. It was showing how you could reprogram cells to restore the function they had when they were younger. In that case, they were changing the type of the cell, but we didn’t want to do that.

Through a series of these dinners, I basically went down this rabbit hole. Sometimes you just can’t get something out of your mind, and you need to keep reading more and more about it and thinking about the implications. This is kind of what happened to me with Bitcoin in 2010 or 2011.

I started to feel about epigenetic programming the way I did about Bitcoin at that time. I just couldn’t stop thinking about it for 3 months. When that happens, you want to hold on to that moment of inspiration or motivation and try to do something with it before it fades, because it’s perishable.

Through a series of these dinners, we happened to find the right group of people who came together. That turned out to be Jacob Kimmel, who’s now the CEO of NewLimit; Blake Byers, one of the other co-founders; and Greg Johnson, who co-founded the company. I put some initial money in, committed to that, and helped them with some of the company-building part of it, but the science was totally new to me. I was basically just a sponge of knowledge, trying to soak up as much of it as I could.

Long story short, the science progressed faster than I thought. We’ve got a lab now in South San Francisco with about 50 or 60 people running these large-scale experiments. It starts with AI, as with all things these days. You can test millions of hypotheses about what sets of transcription factors—which are these proteins—could reprogram different cell types.

It’s a massive search space, so we’re using AI for that. We’ve built the leading frontier model for epigenetic reprogramming, which recommends the next set of experiments that we should run in our lab. These are large pooled screens, and we run them at as big a scale as we can.

But it’s like moving atoms and liquids in a lab, not just software. We do these large pooled screens to see if we can get phenotypic hits for compounds that might be able to reprogram these cells. We then run the best ones through functional assays, which are typically animal models, and those are a little more expensive and a little slower. We run the best ones of those through human trials.

The exciting thing is that we’ve now demonstrated successful reprogramming of at least one human cell type in these humanized mouse models. We’re now testing that in non-human primates, and the first phase 1 clinical trial will be launching next year. There’ll be 3 to 10 other drug candidates coming off this platform in the coming years.

That’s been an incredible journey for me, just to learn about what it takes to try to spin up another company, try to see where I can be helpful, and be humble about the things I don’t know. I want to make sure that I’m focused on Coinbase. That is my full-time job. I’m basically just an investor and board member at NewLimit.

I think it’s fun and intellectually stimulating to try to create these companies that can advance the world. It’s hard enough to get any one company to work. I probably tried 10 different ideas before Coinbase started to really work, right? You want to make sure you don’t try to do multiple things too early. That distraction can be really dangerous. It’s arguably irresponsible to try to go do it.

9. Coinbase and the Everything Exchange

But there are some people—I think you’re one of these people—who have managed to help incubate a lot of different companies at this point. Joe Lonsdale has done this. Sam Altman has done this. Palmer Luckey now has a handful of companies, like Anduril. Of course, Elon is sort of the most famous example of this.

I think there’s a blurry line between being the CEO of one company and being a pure investor. In some sense, you’re both capital allocators. At a certain size of company, you’re basically just allocating budgets and capital there. There’s a bit more to it than that, but I think having a small handful of things you’re trying to do can make sense.

You don’t want to get overwhelmed. If a couple of these things have a problem at the same time, you’re just not going to be sleeping and probably ignoring your family and stuff like that, which is not balanced or healthy. But if you’re just doing one thing forever and you have the potential to help incubate more in the world, then you’re limiting your potential.

Yeah, I totally agree. I think more and more founders are waking up to your realization and your actions. Once you hit a certain scale, or a point where things just have escape velocity and they’re working, there’s enormous leverage. Sometimes you’re able to do incremental things, and I think Daniel Ek would be another good example of that.

He’s been doing really interesting things in Europe. He’s made that successful transition to incubating more companies over time and figuring out where he can plug in to help. Some of it is capital, but really his time is probably the most valuable part.

Has NewLimit announced where it’s starting, or what indication areas it’s starting with, in terms of what it’s hoping to impact biologically or from a patient-centric perspective?

Yeah, it has. On the website, NewLimit.com, you can check out Therapeutics. We’ve published the roadmap there. The first 3 cell types that we’ve started running our platform against are liver cells, or hepatocytes; vascular cells; and immunological cells, through your immune system—T cells most specifically.

We’re following a pattern that’s kind of similar to the GLP-1s, like Ozempic and all these things. Initially, in drug design or drug search, you’re going after an indication with a high unmet medical need. It’s sometimes a smaller market, and that’s just how drugs are developed, because you go after things where people have no better options.

For instance, with our liver program, the initial phase 1 clinical trial will be in people with alcoholic liver disease, or ALD. Today, they don’t really have any other options other than a liver transplant, and most of them don’t have that in time. I think the survivability within 12 months once you’re diagnosed is very low.

We’ll do a phase 1 in that, and that alone would be a pretty large market. Sadly, there are a lot of people in that situation. We estimated that, if that drug alone works in ALD, it might be worth $20 billion or something.

But the goal here is not just to make a liver drug for alcohol damage. It’s to ultimately make a drug that healthy people can use. If you’re 40 and you want to have a liver that’s like your 20-year-old liver, or your immune system or your vascular system, or eventually your brain, your muscles—all these different cell types—your skin, that’s a big one, right?

That’s probably a $1 trillion market right there if you can get your skin rejuvenated.

But the idea is that, over time, you’d be able to take a cocktail of these therapies, and they would target different cell types with slightly different payloads. They would all have the same theme of reprogramming your cells to restore the function they had when they were younger.

Most of the diseases that take us out are correlated with age. When you’re in your 20s and you get an errant cancerous cell or a little bit of damage to your liver, your body recovers from it. As you age, your cells lose this function, and most of the diseases that take us out become highly correlated with age.

10. Looking Forward Five Years

Instead of going after each disease and its symptoms, we’re trying to go after the meta-problem: How do we restore the function your cells had when you were younger? Maybe that would be a way to take a bite out of a lot of these major diseases and attack the root cause.

This goes back to the Yamanaka factors that you mentioned, where they were able to change the imprint for both age and cell type, and you’re saying to focus just on the age component of it. You’re reversing the potency of the cell to something that has more regenerative capabilities. Younger cells or tissues generally are able to rejuvenate or regenerate better than older tissues, right?

You see that with wound healing, injuries, and other things like that. Somebody’s in an accident when they’re old versus when they’re young, and they get better much faster. It’s really fascinating stuff. I think you’re this really unique and wonderful mix of somebody who’s grounded and pragmatic, but who also thinks big and goes after these giant shifts or trends. What are you most excited about over the next 5 years, either for Coinbase or NewLimit, or in general—in the world?

I’m always thinking about this in my free time and on my weekends. What are the most important trends? What is technology making available now that wasn’t available before? What do I want the future to look like? How can I help shape and create that for civilizational progress?

I think it’s good for more people to do that. As an aside, it’s not exactly what you asked, but I think everybody should try to contribute in some way to civilizational progress and think about what they could build. It’s actually really hard. It’s easy to be a critic, to poke holes in what other people are doing, or to be an activist or whatever.

That’s kind of a common thing among a certain generation. I think it’s better to try to be a builder. You’re going to fall flat on your face, and you’ll get an appreciation for how hard it is to contribute sometimes. Maybe it gives you more empathy, but it’s much more fulfilling if you help build something in the world that benefits a lot of people.

To answer your question directly, my main focus is going to be on agentic finance. That’s what’s happening at Coinbase, and then stablecoin payments and everything around that. Crypto is really only powering about 0.5% of global GDP right now.

There may be about 700 million people in the world who have used crypto or hold crypto, but not all of them use it every month. The number of monthly users might be more like 50 million to 100 million people. We need to get that to 1 billion or more, and that’s how we’re going to truly update the financial system. We can do that with humans, and we can do it with agents.

That’s going to be a big focus, and then, of course, NewLimit. But if you want to talk about frontier ideas that I’ve been contemplating, one is cognitive enhancement.

There are things people have realized about the development of children. If you give them enough folic acid and certain nutrients in certain countries, it actually raises IQ; they develop better. It might be interesting to host a dinner on this at some point to discuss the various ways someone could target therapies—in adults, children, embryos, or whatever—to level the playing field a little bit in terms of the baseline cognition and IQ that people are born with or could be born with.

I think that’s another one of these big meta-problems. I haven’t found a great thesis about exactly how to tackle it, but if somebody has one, let me know.

That’s super interesting. It’s basically: How do you increase the collective intelligence of humanity?

Yeah, there’s the biological form of that, and then there’s the digital or AI form of that.

Exactly. I really think we’re in a race right now. AI is getting smarter and smarter, and I think humans need to, at some point, take control of our evolution and get smarter and smarter. Of course, there’s the hybrid of the two with brain-machine interfaces, where some people will choose to merge and some people will not.

Even if, as you know, I believe—and as you do—that there are probably going to be more AI agents than humans in the not-too-distant future, humans still need to advance and progress. That’s important. I think there’s a lot that we can do with these AI tools to benefit humanity: healthier outcomes, living longer if you want, having more energy, less back pain, waking up not tired in the morning, and increasing our cognition.

We can have healthier offspring and children who are free of diseases. There’s a company I invested in called Preventive that’s doing embryo editing, which is a controversial topic. People have seen movies like Gattaca and things like that, but if you look at the pure research, 80% of Americans support embryo editing for disease prevention.

Nobody wants these rare genetic diseases for their children. Everybody wants healthier, better lives for their children. I think that’s one of those topics that seems controversial today. I think it’ll become more normal over time, and at some point it’ll be abnormal to have children without doing this kind of screening and editing, because it’d be like, “What? This is so risky. Why would you do that?” It’s like driving without a seat belt or something like that.

I’ve always wondered whether, in a future where gene editing is reasonably common, there may be packages that you buy. There’s the intelligence package and the health package. Overall, if you think about different ways to be intelligent, there’s processing speed, recall, and five or six different ways you can imagine different forms of intelligence.

I sometimes wonder whether these packages will effectively diminish the neurodiversity of humanity because they’ll push everybody toward the same package. It’s almost like you’re buying the Ford Model T of intelligence for everyone or something. I’ve always been curious about the implications of active editing relative to human diversity and capability sets, and whether parents will default to the basic package or whatever, which will always be the same thing.

It’s a random aside, but there are all these weird downstream implications of some of these things.

Yeah, I’ve heard that argument before, and I actually think it’s the opposite. I think it could increase diversity. The reason is that there’s unlimited diversity in human preferences, right?

11. Special Economic Zones

If it were this simple—and it may not be this simple—parents would say, “Okay, do you want to have a great artist, a great general, a great engineer, or a great politician?” Every parent is going to have some different favorite thing. By the way, that assumes that they’re mutually exclusive. A lot of this might just be raising the floor, and then the kid is still who they are and what they were going to be; they just don’t have schizophrenia or depression or something, right?

So, anyway, I actually don’t think it’ll decrease diversity. In the limit case, you could get diversity that’s really out there. You could have kids born with gills who can breathe underwater, or have wings and fly, or something. Who knows how far it could go if you really can take control of your evolution?

One thing that’s long been discussed in crypto circles, libertarian circles, and elsewhere is different forms of special economic zones. Obviously, those have been implemented at scale in some cases. Shenzhen, in China, exists in part as a large-scale manufacturing region because it became a special economic zone. There are others that have been discussed in the Philippines relative to certain U.S. interests.

There are really interesting things there. How do you think about that, or how has your thinking evolved over time, especially since crypto has played a prominent role in some of the underpinnings and philosophies behind these things?

Yeah, I’m super interested in special economic zones because, again, I’m interested in meta-problems that unlock progress across a whole category of things.

And I think if you look at what slows down progress in many parts of the world, it is overregulation. Europe certainly could be an example of this. The U.S. has its own version of it that is not perfect, and there has been interest at various points in different solutions to this. There’s seasteading and Mars and going up, brain upload into cyberspace, and all these kinds of things.

There’s one called Próspera, which is in Honduras. Coinbase actually made an investment in that just to learn more about it and see what was going on. They’ve had a mix-up, I’d say, with the local government there, but they’re getting closer to a resolution, is my understanding, and I think they’re going to make it. I think they’re going to do a good job.

There was a little bit more interest in this before the most recent Trump administration, just because the deregulatory environment has created a lot of opportunities here in the U.S., and that’s a great substrate market to build on. There was actually an idea that was discussed at one point, I think in the campaign, about something called Freedom Cities. If people Google it, they can see that idea. The idea was to try to create something on federal land in the U.S. The federal government owns lots of different pockets of land.

And if we say, “All right, we think it’s important strategically for America to get really good at nuclear power or drug discovery or having crypto or drone delivery or whatever it is,” you could take a piece of land somewhere. The U.S. has a ton of land, right? You could take a piece of this federal land—10 or 100 square miles—and say, “You know what? In this area, you don’t need to go through all the normal red tape on the EPA and all these things, or the local community. You can just build data centers there. You can build nuclear power plants there. You can do unlimited drone delivery there. You don’t have to get approval from the FAA.”

Because we want to create these zones where experimentation can happen in a safe, contained way to see what works and what doesn’t. If something works via experimentation, then you can go through the normal process in the rest of the U.S. to see if it’s ready to be brought to that location. The problem right now is that, in many cases, there is no sandbox to go try this innovation, and that is incredibly stifling. You can’t really solve a problem as an entrepreneur unless you can iterate on it, right?

12. Conclusion

Anyway, I think it would be really fun to get some of these purpose-built cities or special economic zones in the U.S. As you mentioned, Shenzhen, Singapore, and Dubai have been very successful in many other parts of the world and driven a ton of economic growth. In some ways, the U.S. is 50 special economic zones because of federalism and the state system. That level of competition has, I think, allowed the U.S. to be the leader globally in this. It’s just that we could go even further and do even more.

Yeah. Very exciting. Brian, thank you so much for joining me today.

Of course. Thanks for having me.