The Future of Crypto: Stablecoins, AI Integration, and the Path to Mass Adoption (Jesse Walden, Founder & Managing Partner at Variant)

Speaker A: AI is a technology of abundance. You can create anything you can possibly imagine with AI, and the result is just an explosion of new information, new creativity in the world. And crypto, on the other hand, is the technology that is fundamentally about scarcity. Scarece digital assets. The ability to own digital property is built on top of this idea of scarcity. Speaker B: You're tweeting about what the next NFT marketplace should look like and asking for suggestions. Where are we now? What needs to happen for this to become more useful or more important than it is?

Speaker A: I think NFTs are still incredibly promising. The number of actual non-fungible tokens on blockchains is going to grow many, many orders of magnitude over the next 10 years. What we've seen is there's less interest in consumer speculative side of that, and hopefully there's going to be more interest going forward in developer opportunity. And the new opportunity is just build a product people want that utilizes ownership as a keystone of the user experience. Speaker B: Hey, I'm Mario, and this is The Generalist Podcast. As the saying goes, the future is already here.

It's just not evenly distributed yet. Each week I sit down with the founders, investors, and thinkers who are already living in that future to help you see what's coming next and understand it more clearly. Today I'm speaking with Jesse Walden, the founder and managing partner of Variant, an early-stage crypto-focused venture capital firm. In my opinion, Jesse is one of the industry's clearest and most differentiated thinkers, something he's exhibited across multiple cycles. In our conversation, we discuss where crypto is today, the biggest strides the industry has made since 2022, where Web3 still lags behind Web2, which teams are best positioned to win, and the intersection of crypto and AI.

You'll learn about the state of stablecoins, why the NFT trough of disillusionment won't last, and what to expect from DAOs 2.0. I walked away from this conversation with a crisper sense of where crypto stands, the catalysts driving the current cycle, and where we might be headed next. Jesse's ability to see and articulate the technical primitives, economic forces, and super narratives at play make this essential listening for anyone interested in understanding the industry right now. This is a new podcast, so if you like it, I hope you'll consider subscribing and joining us for some of the incredible episodes we have coming up.

Now, here's my conversation with Jesse Powell. Jesse Walton. Jesse, thank you so much for being here. Uh, maybe to, to kick things off, we could start with a little introduction about you and, and your work at Variant. How would you describe, uh, what Variant does and, and your sort of role within it? Speaker A: Sure. So yeah, I'm the founder and managing partner of Variant, and we are an early-stage crypto-focused venture capital firm. So that means we invest in founders at the earliest possible stage who are building in and around the crypto and blockchain space.

Speaker B: Awesome. Uh, you know, I think one of the reasons I've been so excited to chat with you is you consistently have some of the most interesting and, um, differentiated ideas that I come across in crypto. And that's been the case now through, uh, at least a couple cycles. Um, one of the, the ways that I think you define your work at Varian, which I thought was really interesting, is, uh, that you believe all of the world's value will one day be tokenized and on-chain. Why is that something you believe, and how have you come to to that belief?

Speaker A: I would answer this by way of analogy, which is, which is always imperfect, but I think that tokens are very much akin to packets, packets of data. Packets are the kind of core unit of the internet protocol. And, and so if you, if you kind of take the really zoomed out view, what the internet's done over the last 20, 30 years is it's turned all the world's information into packets. Packets are this kind of empty, empty container that can move any type of information across the internet. You know, as a result of the internet's growth, we, we figured out ways to program all the world's information to fit these packets and move them instantly anywhere in the world.

I think a good proxy for thinking about crypto tokens is to think of them as sort of like packets for value. They are similarly these empty containers. You can program them with any kind of value. And blockchains enable you to move tokens instantly anywhere in the world. And, and so, um, you know, to tie it together, just as we turned all the world's information into packets, I think we're going to turn all the world's value into tokens, and we're in the very early innings of that. Speaker B: And just to sort of draw that thread even, even more fully for folks who are maybe sort of thinking about some of this stuff for the first time, like, why is that the superior way to transfer value versus some of these other data packets from the past?

What is the sort of, yeah, obvious advantages that you think, okay, this is clearly the way that things end up? Speaker A: Right. So, you know, the, the reason that, um, all the world's information is, is packets is because the internet protocol is this open permissionless protocol. And, you know, that the entire world now speaks that protocol. The world's, you know, machines speak that protocol. Um, so the open permissionless nature of it is what caused it to be globally adopted. It's a standard, right? And, and, and as this foundational standard, you know, everyone could use it without fear of it, you know, getting changed or charging them rent.

Public blockchains are, are very similar. They're open, permissionless networks. Anyone can build on top of them, and they have very strong guarantees about their security, their reliability. Um, and so tokens running on top of these permissionless public blockchain networks, um, are, are very much like the internet. Protocol itself. Tokens are a standard that anyone can use, and, and, you know, this is the reason that tokens are becoming more and more pervasive and increasingly, you know, capturing more of the world's value. Whereas, you know, prior standards are proprietary, they're, you know, managed by nation-states, by companies, they're not interoperable, they're difficult to build on top of.

So it's the open, permissionless nature that led to the internet success. Similarly, you know, it's what I believe will lead to, you know, public blockchain success and success of tokens, you know, capturing all the world's value. Speaker B: Amazing. Well, uh, to sort of take a step back and, and think about where we are a little bit, I don't know what you would say. Are we the start of a new cycle, midway through a new cycle? Uh, how, how do you think about this moment in crypto? Speaker A: Yeah, I, it's, you know, I'm, I'm, I'm, I would say I'm not, um, I'm not a macro economist, but it does seem like, you know, every, every 4 5 years, we get some kind of observable pattern in certainly in crypto markets and I think markets more broadly.

So exactly where we are, I couldn't tell you. If you take a really kind of zoomed out view of crypto, I think it's been very steady progress over the last 15 years, which is about how, you know, roughly how old the entire industry is. So within, you know, the 4-year, 5-year view, there's been a lot of ups and downs, a lot of volatility. Um, but from the zoomed-out view, it's a very steady up and to the right march of progress. Yes. Speaker B: And, um, speaking of those sort of moments of volatility, obviously, you know, 2022 was when sort of we had our, you know, beginning of a new winter a little bit.

Over the years since, there's been sort of, as you, as you mentioned, this sort of march back upwards. And, and, you know, sort of keeping with that trend of, of up and to the right, when you look back at, you know, where we are today and, and when things really sort of for the broader view fell apart last time. Like, what are the main things that are different in crypto 2025 versus crypto 2022 in your view? Like, what did we learn from this cycle? What did we get out of it?

Uh, yeah, what can we sort of point to as the main points of progress in your view? Speaker A: Well, yeah, there, there, there's been a lot. I mean, I'll start with the, the most zoomed out view, which is, as I was just saying, you know, over the past 15 years, it's been this, this steady march up and to the right. And within that 15-year span, there's many of these kind of fractal up and down cycles. And I think what you can kind of look at each of these cycles as is a kind of microcosm of Carlota Perez innovation hype cycle where, you know, you start with this kind of invention, there's this explosion, a lot of speculative frenzy around this new idea.

That speculation leads to a lot of investment, and very often the investment is productive in some ways in that it funds infrastructure, funds build-out. But of course, the, you know, expectations get ahead of the fundamentals, and eventually there's a correction. And during that correction, a lot of people within this space find themselves in this sort of trough of disillusionment. But there was a core there, there usually. Um, and out the other side, um, you, you find yourself on this steady slope of productivity. That's the Carlota Perez hype cycle. Um, and I think that you could look at the last, the, the last market cycle, which is probably the most pronounced, um, to date, 2021 to 2022.

We saw the, um, kind of inflated expectations and, and speculative frenzy. And, and then on the other side of that, FTX, there was this kind of trough of disillusionment for the better part of '22, '23, and '24. There's been a lot of progress throughout that period. Um, the, the, the high level is what I would say is a lot of things in the space are well on their way to productization, which I would say is, you know, on their way up the steady slope of productivity. These are things that were invented, you know, 10, 15 years ago.

So case in point, Bitcoin, right? The, the, the granddaddy of, of the space. It was invented in 2009. Today, I think safe to say it is, you know, it is being productized as a store of value. Institutions, governments are, you know, are buying this product as a store of value. Stablecoins are another example where, you know, stablecoins have been around now roughly 10 years, but they're now sort of hitting a scale and velocity of growth that I think really signals they have very strong product market fit. They are being productized.

I think DeFi is also on this steady slope of productivity. And so that is, I would say, the thing that is probably most different is there's a number of things in, in the space now that are on this steady slope. They're being productized, work, and they're no longer in the invention phase. Simultaneously, there's a bunch of new things that are, are kind of more early in this, you know, trajectory and are still in the invention phase. Some of them are probably overhyped. Um, but it's that sort of continuous cycle that, that's playing out at a micro or fractal level within the space consistently.

And there's now a lot of things on this, on this march of productivity. Speaker B: Stablecoins is, you know, I think such an interesting one. I think people come to crypto from lots of different angles and, you know, probably everyone gets into it in some respect because of Bitcoin and, and ETH and stuff like that. But, uh, you know, beyond those two, for me, I would say the thing that really got me excited in, in some of the earlier days was the stablecoin idea, the idea that, you know, you can sort of, especially across emerging markets, maintain, uh, some degree of, you know, security around the wealth you earn and protect against inflation and all these sort of things.

And it has been so interesting to see how it's developed recently. You had some really interesting writing about stablecoins where you talked about the idea that they are both a consumer convenience and a messaging protocol. And I thought that was like a really interesting way of talking about this thing that, you know, at least at one level is, is very simple. It's, you know, a dollar digitally in some respect. What, what is the sort of like the two pieces of that for you? And like, what does that say about the advantages of stablecoins?

Speaker A: I, I realized I actually didn't fully answer your prior question, which was sort of what, what's different now, right, in cycle. And, and asking, you know, digging, double-clicking on stablecoins just made me realize I should have address this because it's probably the single biggest difference, uh, beyond what I was just describing. And that is like the regulatory environment has completely changed in crypto, um, really since the election, um, in November. And today we're a couple weeks into the Genius Act, you know, passing. Um, and the Genius Act is the first crypto legislation in the United States.

It's also the most bipartisan legislation to pass in the United States, I think, in the better part of a decade, which is, you know, to say that it's not just the new administration, both sides of the aisle are excited about crypto and specifically about stablecoins. And so I think that is, you know, you can't understate how big a deal that is, or you can't overstate how big a deal that is. Really, the entire 15-year cycle that I was just describing of invention to productization happened, you know, in, uh, under regulatory uncertainty.

And, and for a lot of it, not just uncertainty, but like an adversarial regulatory environment. Speaker B: Can you give folks like the quick sort of précis on, on Genius and, and what it really means? Speaker A: Yeah, I mean, at a very high level, it, it, it sort of just sets clear rules of the road for how stablecoins are to be issued and managed. And, and so you can think of it sort of belonging to the class of financial regulation that, you know, keeps banks, or is intended to keep banks safe and on the right path, right?

It's, you You want banks to be, you know, have a regulator, you want them to be audited and have certain reserves. And, you know, the Genius Act is very similar, but for stablecoins, it sort of describes who can be an issuer, you know, how to qualify, what type of reserves are acceptable, and so on. And so the result of this legislation, we think, is that, you know, you're going to see this productization of stablecoins really accelerate. As banks and the, the whole kind of traditional financial markets start to adopt them at scale.

Speaker B: Amazing. And, and so, yeah, to pull us back, this regulatory change is one of the, the big things that we did not have in 2022, uh, and, and the sort of level of certainty around it, right? Speaker A: In fact, we had, we had the exact opposite. We had, um, kind of no, no regulation. We had kind of an adversarial regulatory environment in the United States, and the result was a lot of founders going offshore where there was even less oversight. And, and that ended up causing a lot of the problems that, you know, that set the space back in 2022.

Speaker B: And so to go back to this notion of sort of stablecoins having these, this dual purpose of, of consumer convenience and, and messaging, uh, protocol, like, yeah, what, how did you sort of come up with that framing and what does it tell us about the tech? Speaker B: And so to go back to this notion of sort of stablecoins having these, this dual purpose of, of consumer convenience and, and messaging, uh, protocol, like, yeah, what, how did you sort of come up with that framing and what does it tell us about the tech?

Speaker A: Sure. So, so first off, starting with the messaging protocol side of it, I think, you know, again, tokens are this kind of standard, like packets. It's, it's, it's, you know, it isn't open-source standard, um, and, um, sending stablecoins across blockchains, you know, you can think of this as, as a message, a new messaging layer similar to SWIFT or ACH. These are messaging protocols. Different about stablecoins, um, compared with these legacy messaging protocols is that, um, with stablecoins, the, the medium is the message. And, and what I mean by that is the legacy messaging protocols for payments, SWIFT and, um, ACH, they, you know, they, they— you send a message and it's up to the bank on the receiving end of the message to update their internal ledger.

With stablecoins, there's one global ledger. And, um, you know, when you send me a dollar, um, I have one more. Um, and it's a bearer instrument. What that means is, you know, the message itself contains the payload, which is the token. And, um, as the holder of that token, I've got it, I can spend it immediately. If I'm a developer, I can program it, right? So it's a lot more flexible than legacy protocols where you're dependent on this third party at the end of the message to kind of update their internal system.

That's the key difference in the messaging protocol part of it is stablecoins and all crypto tokens are bearer instruments, meaning whomever receives the message essentially has the asset and do with it what they will. Now, um, in terms of consumer benefit, um, I touched on this a little bit, uh, you know, stablecoins being sent over the— over this new rail, block— public blockchains— faster, cheaper, global, and more programmable. Those are the consumer benefits. Programmability, I think, is a huge one, um, because it unleashes developers to build experiences for consumers that you know, are much better than what the legacy systems can offer them.

Today we're seeing a lot of those benefits— fast, cheap, global, programmable— show up and, and sort of get widely adopted for, for things like, um, cross-border payments or dollar-denominated savings products. And these are things that are, you know, primarily of value outside of the United States, um, where access is, you know, is, is challenging, or cost is, is challenging. You know, a lot of the convenience today is being realized out outside of the US. Here in the US, stablecoins are not yet convenient for two reasons, um, but I think they both have a solution.

So the first is they suffer from a classic sort of network effect problem, right, um, where, you know, if no merchants are accepting stablecoins, they're not convenient to spend, right? Um, but I'm really excited about the potential that of using and specifically programming incentives to bootstrap payment networks that are denominated in stables. So again, the cool, the cool thing about stablecoins is they're programmable. Developers can kind of do anything you want with them. Stablecoins could potentially be earning yield in DeFi. They can be earning a yield from, you know, the treasuries that are backing them.

So I do think in time we'll start to see some of those kind of incentives be used to bootstrap the network effect that make them convenient for payments even, even here in the US. The, the second problem, or second inconvenience in the US, is on and off ramping. And this is really true globally, but again, the, the other benefits of them being faster, cheaper, and global out, you know, in the rest of the world offset this inconvenience. But in the United States, you know, we all know it's a pain to on and off board to crypto, including stablecoins.

It's kind of a UX nightmare. You got to log into Coinbase or an exchange and move money in, wait, move money to your wallet. This is, I think, one thing that the Genius Act, uh, addresses head-on. So, um, again, anyone can become an issuer, anyone can start to touch, um, stablecoins. And so I think you're going to start to see banks custody stablecoins, you know, enable their users to access them much, much more easily than they can today. So anyway, that, that, that's sort of a rundown on, on, you know, the convenience of, of stablecoins, like where it exists, where, where there's still room to run.

And I think all this is— all the issues today are going to get worked out in the next, you know, 2 to 5 years, and that's when you're going to really see them hit escape velocity. Speaker B: With Variant, how have you ended up sort of playing the stablecoin opportunity? Like, have you guys sort of been a part of any of the stablecoin neobanks, or more on the infrastructure side, or is it still the case that, like, actually maybe the, the one that you're really looking for hasn't, hasn't arrived yet, and it's these sort of net new networks that you're looking for?

Speaker A: All of the above, actually. Speaker B: Okay. Speaker A: So we're very excited about stablecoins. What I would say is, you know, we've been investing in what we know best, which is our sort of crypto-native opportunities. And, you know, so I think directly downstream of the growth of stablecoins is growth in underlying infrastructure, right? It's very cheap to send a dollar, but it's not free, right? And it, you know, so at scale, we think stablecoins, and they already are, are going to generate a lot of transaction fees for the blockchains on which they move.

And of course, we're invested in many of the leading blockchains where stablecoins are, including Ethereum. We also think that directly downstream of stablecoin growth is growth in DeFi. We're investors in decentralized exchange protocols, decentralized lending protocols, which are some of the venues where you're seeing stablecoins really show up and grow at scale. And we think that's going to continue. One of the companies in our portfolio, or projects rather, is Morpho. Morpho has this integration with, with Coinbase where you can borrow dollars against crypto. So if you've got some Bitcoin, you can go, you know, go on Coinbase, borrow dollars against it.

Those dollars you're borrowing are stablecoins. Um, and, um, you know, couple months into this partnership, they've got, um, you know, over a billion dollars of Bitcoin being used as collateral and, and, you know, roughly 50% of that coming out in, in dollar-denominated stablecoins. So we think that would, again, with Genius, that's really gonna accelerate. And then, yeah, we're also investing in kind of stablecoin pure plays. Genius, we think, is going to result in the issuance of thousands and thousands of stablecoins. And again, they're being deployed over multiple chains. Fragmentation becomes a new problem to solve.

And we made one investment there in a company that's trying to solve that fragmentation. And I would say increasingly, we're looking for founders who are bringing the kind of consumer convenience to end customers. We typically invest in the Western Hemisphere, um, and as I was just describing, we think there's still kind of friction that needs to get erased, and, and that's going to happen over the next 2 to 5 years. So we, we think there's opportunity to, to come at the consumer level too. Speaker B: We talked a little bit about, you know, Genius and, and some of this regulatory shift, uh, more recently.

One thing that I've been thinking about, you know, in the new sort of political environment is, uh, is it possible that this gift gets curdled somehow? Like, is, is this, uh, you know, new regime so positive on crypto that we sort of push it too far and, you know, the abuses that can happen in this system get so exacerbated that it actually sort of, you know, in the fullness of time creates an even bigger backlash next time? Like, is that something that you worry about at all? Speaker A: I would say I'm, I'm more optimistic than I've ever been about the, the being like getting it right, basically.

Um, what I will say is it's still the case right now that there's, um, you know, there's a few milestones in terms of policy that we have not hit yet. So Genius was, was the first, and, and, you know, frankly, the biggest one, you know, biggest success, uh, in, in terms of policymaking the industry's had. But there's a number of other irons in the fire, namely, you know, market structure, which is currently going back and forth between the House and Senate. There's also rulemaking going on at the agencies, the SEC, the CFTC, and so on.

There's also a really high-profile case being heard in federal court right now concerning Tornado Cash, which really has implications for the whole crypto space. And that's to do with whether developers are liable for writing open-source software. So there's a number of these kind of things in flight right now that need to land in order for, um, you know, crypto to succeed, and, and specifically for, for good actors in crypto to succeed. The lack of regulatory clarity over the last, you know, 10+ years is what has allowed bad actors to succeed, and the bad actors are the ones that, um, have created all the problems and setbacks for crypto.

So we're still, I would say, in the thick of it. We're kind of in the fog of war where you know, even the, even the president himself is, you know, launching a meme coin and, and, and doing things that, um, you know, a lot of folks in the industry like aren't that excited about, frankly. Yes. Um, at the same time, you know, it's, it, it's the current administration that is working on all four of these kind of policy legs. Um, and that's very encouraging, right? So I would say like the key to avoiding the scenario you described where we do have kind of another massive setback is landing the plane on, on those policy initiatives, getting the rulemaking right, getting market structure legislation done.

Um, because that, that's what's going to enable the good actors that kind of keep us on the right side of things. Speaker B: Yeah, you mentioned, you know, the meme coin story, and I think, uh, you know, when you look at the last cycle of crypto, you have stablecoin, uh, productization, you have so much institutional interest with, you know, the ETFs, the regulatory changes, all these things. And then, you know, I think the most sort of consumery part of it has been this meme coin craze, which from my vantage, and I'm sort of always just keen to be open-minded, so maybe you'll push back and take us in a different direction.

To me, I'm like, this is literally the last possible thing crypto needed based on the cycle beforehand. When you look at it, is there anything about it that you're like, maybe there's a positive externality here? That, you know, uh, we do get something productive out of this? Or do you just see it sort of like in keeping with a lot of the, you know, behavior that has, has, has been the, the part of this industry that's really been holding itself back in many ways? Speaker A: Yeah, it's, it's a, it's a nuanced question.

I got a lot of thoughts here. I, I would say, you know, there's two things that have been working in crypto indisputably over the last couple of years, um, and it's— and those two things are on a barbell. They are like polar opposites of one another, and they are speculation and stablecoins. Speaker B: I'm all in on that. I love that. Speaker A: Always worth a dollar, non-speculative. Like, that's, that's the purpose of a stablecoin, and it's working. On the other hand, you know, speculation— you know, Bitcoin is the original speculative asset in crypto, um, and, and there's been a number of other speculative experiments along the way, including meme coins.

And meme coins really hit a stride over the last couple years. Meme coins have been around 10+ years, you know, Dogecoin being the biggest, right? Um, but, you know, they, they really hit escape velocity over the last couple years. And I think, you know, the fastest company not just in crypto but to $2 billion revenue is PumpFund, which is, is a meme coin launchpad. Um, the reason these two things are working, um, on, on both ends of the spectrum is the same, which is that, you know, the infrastructure that enables both has finally gotten good enough.

So stablecoins, fast, cheap, global, is a result of the infrastructure improving. Similarly, you know, going from Dogecoin, which was a fork of Bitcoin that was very technically complicated to pull off, um, you know, to, you know, anyone being able to upload an image and launch a coin in seconds is also a result of fast, cheap infrastructure. And so in a way, I would say I'm not surprised, right, that, you know, that stablecoins were, were kind of one of the dominant games in town, um, because it's a result of this kind of fundamental progress on the infrastructure layer.

And, you know, both stablecoins and meme coins are ideas that have been around for 10 years, and they just got, you know, to their final form, or, or, you know, somewhere closer to it. Double-clicking on meme coins specifically, um, you know, I think there's no question that, um, you know, there, there's a casino element to it. I think another way to look at meme coins, um, is as a, a kind of a weighing machine for something that's very valuable, uh, on, on the internet, and that is attention, right? So attention on the internet, we know, is very valuable because it is monetized typically with ads, and, you know, social networks, um, monetize it very, very effectively.

I think meme coins are a market that, that, you know, quantifies the value of attention on the internet directly. So if you look at the chart, the price chart of a meme coin, right, very often it's spiky, it's up into the right, and then it's kind of straight down. That— the shape of that chart looks very similar to the shape of, um, the chart of virality of a tweet, for example, or social media post. You know, when something's going viral, it spikes up, and then, you know, very quickly it it goes away.

And so for that reason, you know, I think meme coins as an asset class are not going away, right? It is, um, you know, it is a, it is a market that, you know, prescribes or puts a value on something that is very, very valuable on the internet, and that's going to continue to be very valuable, which is, which is attention. And, you know, there's a lot of experiments going on with, uh, with, with ways to enable creators to capture some of that value in the market. And, and so for example, you know, if you create a meme, um, on Pump or on Zora, um, uh, or, or, you know, with a protocol like Doppler, which is one of our investments, um, creators can capture some of the vol— you know, the value that's being traded back and forth in these markets as a way to monetize their work directly.

So I, I think it's, um, it's not all bad. It, it certainly is consistent with a a secular trend, which is, you know, more financialization, more markets in everything. And, and I don't think it's going away. I think we'll, I think we'll look out 10 years from now and we'll, we'll see that the market cap of meme coins as an asset class, not, not any single one, but as a total kind of asset class, will grow many orders of magnitude from, from here. Speaker A: Always worth a dollar, non-speculative. Like, that's, that's the purpose of a stablecoin, and it's working.

On the other hand, you know, speculation— you know, Bitcoin is the original speculative asset in crypto, um, and, and there's been a number of other speculative experiments along the way, including meme coins. And meme coins really hit a stride over the last couple years. Meme coins have been around 10+ years, you know, Dogecoin being the biggest, right? Um, but, you know, they, they really hit escape velocity over the last couple years. And I think, you know, the fastest company not just in crypto but to $2 billion revenue is PumpFund, which is, is a meme coin launchpad.

Um, the reason these two things are working, um, on, on both ends of the spectrum is the same, which is that, you know, the infrastructure that enables both has finally gotten good enough. So stablecoins, fast, cheap, global, is a result of the infrastructure improving. Similarly, you know, going from Dogecoin, which was a fork of Bitcoin that was very technically complicated to pull off, um, you know, to, you know, anyone being able to upload an image and launch a coin in seconds is also a result of fast, cheap infrastructure. And so in a way, I would say I'm not surprised, right, that, you know, that stablecoins were, were kind of one of the dominant games in town, um, because it's a result of this kind of fundamental progress on the infrastructure layer.

And, you know, both stablecoins and meme coins are ideas that have been around for 10 years, and they just got, you know, to their final form, or, or, you know, somewhere closer to it. Double-clicking on meme coins specifically, um, you know, I think there's no question that, um, you know, there, there's a casino element to it. I think another way to look at meme coins, um, is as a, a kind of a weighing machine for something that's very valuable, uh, on, on the internet, and that is attention, right? So attention on the internet, we know, is very valuable because it is monetized typically with ads, and, you know, social networks, um, monetize it very, very effectively.

I think meme coins are a market that, that, you know, quantifies the value of attention on the internet directly. So if you look at the chart, the price chart of a meme coin, right, very often it's spiky, it's up into the right, and then it's kind of straight down. That— the shape of that chart looks very similar to the shape of, um, the chart of virality of a tweet, for example, or social media post. You know, when something's going viral, it spikes up, and then, you know, very quickly it it goes away.

And so for that reason, you know, I think meme coins as an asset class are not going away, right? It is, um, you know, it is a, it is a market that, you know, prescribes or puts a value on something that is very, very valuable on the internet, and that's going to continue to be very valuable, which is, which is attention. And, you know, there's a lot of experiments going on with, uh, with, with ways to enable creators to capture some of that value in the market. And, and so for example, you know, if you create a meme, um, on Pump or on Zora, um, uh, or, or, you know, with a protocol like Doppler, which is one of our investments, um, creators can capture some of the vol— you know, the value that's being traded back and forth in these markets as a way to monetize their work directly.

So I, I think it's, um, it's not all bad. It, it certainly is consistent with a a secular trend, which is, you know, more financialization, more markets in everything. And, and I don't think it's going away. I think we'll, I think we'll look out 10 years from now and we'll, we'll see that the market cap of meme coins as an asset class, not, not any single one, but as a total kind of asset class, will grow many orders of magnitude from, from here. Speaker B: Interesting. Okay, so the sort of best-case scenario sounds like, uh, it's useful in that it sort of does chart this attention cycle, and so like there's some informational value in that.

And then also, you know, maybe it becomes an interesting source for the internet vernacular of memes and the people who are sort of best at creating that vernacular to capture some of the upside in it. Is that right? Is that like a fair encapsulation? Speaker A: That, that's right. And, and I would, you know, maybe suggest broadening the definition from just like memes in the classical sense to, you know, to just kind of media on, you know, media. And, you know, memes are one variety, you know, news is another. I think You know, when, when the Pope died, if you went on the front page of

fun, you know, there were all these memes about the Pope dying. And I heard from somebody that they found out about the Pope dying on Pump. Speaker B: That's a hilarious indication of that person's information diet, right? Speaker A: Right, right. But I think, you know, if Zora is another portfolio company that's building this space, and if you go on the Zora app, you know, it's a social media feed. Not, you know, visual one, not unlike Instagram, but where every, you know, piece of media is also a token and has a market value.

And that market value is kind of a new type of signal in terms of what, you know, people on the internet are valuing and paying attention to. And so you could look at it as just a new kind of way to curate, you know, a feed. It's an input to the algorithm. Zooming out, I think prediction markets are very similar, right? It's like they are markets for information. And there's a class of people speculating on the value of a given outcome, but there's also a lot of information in that market where people are learning about their news through prediction markets.

They're learning about the outcome of election before journalists are predicting it, and more accurately in many cases. And so there's a similar underlying trend where markets are creating new information. That information can be harnessed to produce news. It's broader than just, you know, you know, laughs on the internet. It's, it's, uh, it's, it's all kinds of media, including news, potentially. Speaker B: I suspect this actually happened, like, not just, you know, meme coins around the Pope's death, but probably meme coins for each of the major candidates who were being considered in the conclave.

And, you know, in that way, it does, you know, sort of— you get a peek into people's revealed beliefs, or at least, uh, you know, where they're, they're willing to make a bet on, on the rest of the of the rest of the public. To go in a slightly new direction, you've written about the fact that, you know, this stage of crypto is really less about these net new ideas and more about this productization. And I, you know, we've talked about that now with, with stables and, and meme coins and all these other sort of pieces of it.

When you think about like the teams, uh, and the startups that are like best equipped to win in this new phase, do they look like the same sort of teams that they did when, you know, these use cases were maybe less legible, or is it a different, different style of company? Speaker A: That's a good question. I do think that the shift from invention to productization is a meaningful one in terms of, you know, what the opportunity set is. And there's some teams that can kind of navigate that shift gracefully.

And then there are, there's a lot of new opportunity in the space as well for teams that are more well suited to productize versus invent. You know, I'd say a lot of the opportunity today is more to do with, you know, execution, um, you know, distribution, uh, you know, great user experiences versus like coming up with the fundamental primitives. That's not to say that there's no primitives left to be built. Of course there are. We're continuing to back teams, um, doing that. But, but I do think that the kind of DNA of, of teams that we see coming into this space and where we think there's a lot of opportunity sort of resembles that of more traditional Silicon Valley startups where there's a very, you know, heavy focus on product, on user experience, on sort of growth.

There's both traditional tools in the toolkit that you can bring to bear to execute. And there's also new tools that are crypto native. And so you kind of need to find them, you know, a mix of both the crypto native entrepreneur and, um, you know, the experienced kind of Silicon Valley entrepreneur. And, and so, but that, that there's, we're really excited about, um, about that. We're seeing them come back to the space. And again, I think that's because there's now kind of, you know, progress and real hope that there's going to be a path to building here.

Speaker B: I think one of the things that, you know, has been historically hard for crypto is like, you know, the lack of, of this hyper-performant infrastructure. And there's been so much investment in that over the years that it's gotten better and better. But also, you know, some of these big user experience gaps that again feel like they've, they've closed, but I don't, I think except in maybe a few cases, there's not real parity with Web2. Like, how, how close do you think we are? What's missing? Like, what— you know, on both of the UX and the infrastructure piece, like, how do you sort of benchmark that?

Speaker A: Yeah, so, so we, we've definitely made a lot of progress, but to your point, I don't think we're quite there yet. I'd say we're, we're nearly there. Um, so for example, you know, it used to be the case that if you want to use any crypto app, the first thing you had to do is install a wallet and, uh, go write down this kind of 24-word seed phrase, which was your private key. And if you lost that, you know, you lost everything. And that, that's a pretty terrible user experience.

Today, you know, we have, um, we— there's a number of companies. One, you know, Privy was just acquired by Stripe. Portfolio Turnkey, which provide, you know, private key management, um, and, and wallets as a service, essentially. And, and so the result is you don't have to do— you no longer have to do that complicated 24-word seed phrase. You just sign up with an email and you've got a wallet. And, and You know, downstream of that is every mobile app becoming wallet-aware, having an embedded wallet within it. That's huge progress. That said, that wallet still needs to get funded, um, right?

You still need, you know, money in it to transact on-chain. And as we talked about previously, the on- and off-ramps are, are, are still a key point of, of friction. Um, and I would say that there's a huge drop-off rate in, in, um, you know, signups today because, um, you know, once you get your wallet, you, you have to go through this funding process, and that's still pretty janky. Again, I think that's going to change because of Genius, where you will be able to get dollars from your bank account into stablecoins and, and into a crypto wallet much more easily, um, over the next couple years.

Um, now one of the other things that I think needs, needs to change is that, um, you know, there's an increasing number of blockchains, there's applications on, on different chains, and interoperability between the chains is still kind of a point of friction. I had a friend texting me the other day, you know, how do I get my, my USDC on, you know, on Ethereum, you know, to this application I want to use on Solana. And, you know, it's not easy is the bottom line. I've always thought this is a huge opportunity for wallets to solve.

And we're investors in Phantom, which is one of the biggest wallets in crypto. And they are multi-chain wallet, they're on Solana, they're on Base, and they do enable kind of swapping between them. Um, but I, I still think there's a lot of low-hanging fruit to just make that kind of completely seamless and, and something that the user doesn't have to think about at all. What I would say is, you know, zooming out, we've made a ton of progress. Um, it's, you know, this is— these are really complicated technologies, um, and, and we're, we're pretty close to kind of putting polish on, on, on like the rough edges, but there's still a few rough edges protruding.

Speaker C: This episode is brought to you by Brex. Fred Adler, the influential venture capitalist of the 1970s, was known for displaying decorative pillows in his office that featured a signature business philosophy: "Corporate happiness is positive cash flow." In today's post-SERP environment, Adler's wisdom feels particularly relevant as founders need to make every dollar work harder. That's exactly what Brex Brex delivers. Their modern finance platform was built specifically for startups like yours and designed to help extend your runway when capital efficiency matters most. With Brex, you get global corporate cards with up to 20x higher credit limits and no personal guarantee required.

Their banking solution has no minimums and no transaction fees while letting you earn high yield from day one with same-day liquidity. Best of all, Brex knows you were born to build, not juggle spreadsheets and finance tools. Their AI-powered platform brings cards, banking, expense management, and travel all in one place. It's simple, scalable, and designed to get you back to what you do best: building. More than 30,000 companies, including 1 in 3 US venture-backed startups, trust Brex to help make every dollar count. Speaker B: Toward their mission. Speaker C: Join them at

com/mario. Speaker B: You, you mentioned Fantom, which, uh, I also got to be a small angel investor in, uh, you know, a few years ago. Amazing product, I think. Uh, and I think a good sort of example of maybe what like the, the next in line class of like really big crypto companies maybe look like, you know, we obviously have had the big exchanges like Coinbase and, uh, you know, maybe Kraken goes public this year and so on and so forth, but As you look at that maybe next in line set of, of secular winners, I think you've talked about like who, who's in that batch, you know, alongside the Phantoms and, and so on and so forth.

Speaker B: Toward their mission. Speaker C: Join them at com/mario. Speaker B: You, you mentioned Fantom, which, uh, I also got to be a small angel investor in, uh, you know, a few years ago. Amazing product, I think. Uh, and I think a good sort of example of maybe what like the, the next in line class of like really big crypto companies maybe look like, you know, we obviously have had the big exchanges like Coinbase and, uh, you know, maybe Kraken goes public this year and so on and so forth, but As you look at that maybe next in line set of, of secular winners, I think you've talked about like who, who's in that batch, you know, alongside the Phantoms and, and so on and so forth.

Speaker A: Yeah, that's a, a good question. By the way, I would say Phantom, um, is a really great example of the, the team DNA I was just describing where they're very crypto native, but you know, the founders, Brandon and Francesco, um, you know, both worked at Twitter prior to getting into crypto. So they have this kind of you know, Silicon Valley experience. They've seen what greatness looks like. I think they've built a really excellent organization, um, with a really sort of laser focus on delivering really great product experiences. Speaker B: Yeah, absolutely.

Speaker A: Distribution, distribution. So, um, so yeah, Fantom is certainly one of, of the key secular winners. I think, I think the opportunity for Fantom here is enormous. Simple way to think about Fantom is, is the gateway to, you know, everything that the on-chain world has to offer. So it's very similar in some respects to a browser, right? Like a browser gets you access to the, you know, all the Web2 applications. Phantom gives you access to all the Web3 applications on-chain. And because Web3 is financial, they sit right in the flow of all the value that users are transacting in that world.

So I think, yeah, very excited about Phantom. I think, you know, Uniswap is another, I think, secular winner, decentralized exchange protocol. Increasingly DEXs are, are doing, um, you know, more and more percentage volume relative to their centralized counterparts. Um, and Uniswap is the, the market leader and has been very dominant for, for a very long time. As all the world's value gets tokenized, that value is going to need a venue to trade. I think Uniswap is, you know, going to keep compounding as a secular winner, um, as the, you know, biggest, uh, decentralized exchange.

Morpho is another I mentioned already. Lending, right? Huge, huge category in crypto that's been driven historically by speculation, right? People wanting to borrow against their crypto to buy more crypto. But as we see more stablecoins, you know, come online and, you know, other types of tokenized value stocks, you know, the value of a lending protocol will shift into this less speculative, kind of more traditional case where, where people, people want to borrow against hard assets. So, so today Morpho is, I think, the second or third biggest lending protocol. We think they've got the right kind of market structure and sort of focus on distribution.

For example, the Coinbase integration that, that's already filed to really take it to the next level. As you know, the, there's a secular tailwind behind them of, of stablecoin growth. You know, which we think is going to be many, many orders of magnitude to go. So those are a few. Worldcoin is maybe another one that I would mention where they're really in a category of their own. Worldcoin, of course, started by Sam Altman. What they're building is what they call proof of personhood. So a proof that you are a unique human as opposed to an AI.

You know, they're signing up over 100,000 people every day. Like at the current rate, I think they're set to onboard like, you know, millions of people. Um, you know, this, this year they've, they've already got, um, north of 30 million, um, installs. And, um, and the goal is to provide a kind of new form of identity for every human on the planet. They're in a category of their own, and, and I, I would say they are increasingly becoming a secular winner in terms of just the distribution that they've got. And that's another great example of a team with real Silicon Valley DNA, focus on distribution, Um, while also understanding the crypto-native opportunity.

Speaker B: Worldcoin's, you know, a good segue into the sort of intersection between crypto and AI, I think. Um, when I think a little bit about the teams that, that we're talking about that, that maybe do have that more traditional Silicon Valley DNA, one of the things that I could imagine being a little bit of an impediment to the next wave of crypto startups is that crypto now has to really compete, uh, with AI. On the opportunity set. Like if you're a really talented builder today, you could decide, hey, I want to build something, you know, around stablecoins and all of that.

Or I could say, you know, this, this large language model revolution or any, any other sort of associated part of this renaissance is, is more fertile and is better for me to, to spend my time on. How do you feel like crypto is faring from a talent perspective? And, you know, is it, has it struggled over the past few years? Speaker A: Yeah. So, so no, no denying the fact that, you know, ChatGPT was 2023, right? And that was a huge inflection point. Speaker B: I think even '22, right? Speaker A: Yeah.

Yeah. Sorry. Yeah, you're right. I think '22. And then, um, yeah, I think by '23 they had like the fastest growth of any, any, you know, consumer product or something. So there's no denying that, um, you know, Silicon Valley loves to chase kind of shiny new thing and ChatGPT was, was it. And so a lot of talent, um, over '22, '23, and the better part of '24 was going into AI. And I think, right, you know, if you think about it, makes sense, right? Like, meanwhile, in crypto, even the, even the best entrepreneurs, like, you know, Uniswap and all these companies I was just mentioning, were like under investigation for 4 years, right?

So like, even the best projects, the best founders, um, you know, the headlines about them were extremely negative. And, and if you're kind of college kids studying computer science, like, you know, you got to make a choice. Are you going to go to where there's all this exciting, um, you know, opportunity, or you're going to go where the government's trying to run you out of the country? So there was definitely, um, a lot of reversal in inflows. That said, you have talent into crypto over the last few years. Um, that said, like, I think Electric Capital did this data analysis where if you take out the spike in talent that came into the space in '21, '22 and stayed for under a year, those would be described as kind of tourists, tourist entrepreneurs.

If you take that out, the compound annual growth rate of developers in crypto has been growing at a very steady clip. I forget exactly, I think it was roughly like 20-25% year-on-year growth. And so I don't want to give the impression that like, you know, talent is not coming in. Because it has been very steadily coming in. It was just this crazy spike during the speculative boom and bust. But if you normalize, you know, over a wider timeframe, it's, it's been compounding. Um, and, and what I will say is along with the shift in the regulatory environment, we've seen a meaningful shift in, in the, the number and quality of entrepreneurs coming into the space now.

Um, and, and that's because I think you're starting to see some of the secular winners you know, double down, win bigger, you know, have the, you know, their investigations dropped, right? I think talent that's been on the sidelines is looking at that saying, hey, you know, this has become a kind of contrarian thing, but there's actually a lot of opportunity here. And we're starting to see that result in more entrepreneurs getting off the ground. And in many cases, it's second-time entrepreneurs, people who've been here for a long time who are finding, you know, greenfield, new opportunity ideas they wouldn't have executed on a year ago and that they're they're coming back to do now.

Hmm. Speaker B: Yeah, that's interesting, uh, normalizing that big spike, because that's also just anecdotally something I'm sure you saw a ton, uh, in your, your seat as a VC, of just like the number of crypto-to-AI pivots, uh, in, in 2022, 2023, um, where you had— you did have a lot of those folks who I think were dipping their toe in, and then the water got really, really cold, and, um, they, uh, for better and worse, decided You know, their talents were best devoted elsewhere. Uh, but that's interesting. Okay. So, you know, things are, are sort of trending in a positive direction when you, when you sort of look at it from a little bit more of a zoomed out view.

Speaker A: One thing I would share is like in, in our portfolio at Variant, you know, we, we had a handful, like, you know, maybe 2, 3, you know, certainly less than, I want to say even less than 5% of our portfolio. Pivoted from crypto to AI. If I had to attribute that, I think there's been much higher, you know, rates of attrition than what we've seen in our portfolio. And that's because I think one of the things we screen for at Variant is we really want to invest in founders who are doing their life's work in crypto.

We invest at the earliest possible stage. So like very often we assume that the idea that we're being pitched is going to change and evolve with the market. So the thing that we really index on is, you know, is this person here to do their life's work? Even if the idea changes, are they going to stick with the general kind of scope of, or intent of the, you know, the thing that they're pitching? And so most of our founders have stayed in the crypto space through thick and thin. And like, we really want to be the best partner for entrepreneurs who are going to be here for the ups and downs.

And so that's why I think we've seen very low attrition in our own portfolio. Speaker B: Well, I can't help but dig into that a little bit, but it doesn't surprise me in some respect because I do think that, uh, well, crypto in general is just such a high variance industry. I find that you meet some of the very, very smartest people you'll ever meet, and then a lot of the folks who maybe are, are not, you know, are balancing the other end of the spectrum. And same for, you know, you meet some of the most principled people possible who have really thought super deeply about, you know, their morals and their values.

And then, you know, obviously you have the total other end of the spectrum. And on the founder side, I think you do have, you know, those, those tourists maybe on the one end, but then you have people who will truly walk through any possible brick wall because of their devotion to this technology and what it sort of signifies for them. What are your ways for figuring out if someone is doing their life's work? Like, have you, have you found any ways that you're sort of able to, to get a feel for that in the moment and as part of a process?

Speaker A: So what I will say is, like, I, I think within the first 5 minutes of talking to someone, generally you, you know, um, which side of the spectrum they're on. Um, and very often it comes out when, when you hear the story as to why they got into crypto in the first place. Um, and as you said, for, for a lot of people, um, it's some kind of principle about the technology that, you know, is really close to their heart or their motivation for, you know, working on the thing that they're building.

And, you know, there's a lot— the great thing about crypto is it's such a multidisciplinary space, you know, combines, you know, deep technology with economics, with, you know, you know, I would say kind of social norms and even politics, right? Because the politics of how to govern these permissionless systems Um, and, and so there's, there's a number of different reasons you might have gotten into crypto. Maybe, you know, you grew up in a country that didn't have a stable currency. Yeah, capital controls. Um, maybe you were just really fascinated by, um, distributed systems where, you know, the control is with the edges of the network as opposed to the center.

There's a number of different fundamental reasons why people get really intensely nerd sniped by this technology. And I'm not saying like, you know, you have to have one of those as a like raison d'être to be in this space, but very often, you know, that comes out at the outset and you can just tell, okay, this person is here to do their life's work. And I would say during that blip in '21, '22, when there were a lot of tourists that were here for under a year, you didn't hear that.

You kind of heard the opposite, right? Which is like you heard some story around how there was, you know, an opportunity to just make a lot of money fast. And, and that's always been another facet in crypto. And so I like to think that one thing we're, we're decently good at, and that we haven't been involved in any of the kind of blowups, or is, is just like sniffing that outlet. Why are you here? Um, and are you going to continue to be here when, when things get tough? Speaker B: Yeah, absolutely.

Um, to return to AI a little bit, you know, I think, uh, on the one hand there is just this like collision of buzzwords that's happening around some of, some of these projects where you just feel like there's not a clear reason why AI and crypto should go together. But on the other hand, there are some really, I think, promising and interesting ideas. I think, you know, uh, the idea that we need to be able to prove our humanity to digital systems is a really, really powerful one. Where have the sort of collisions, uh, that you've spotted that have been most interesting?

Um, yeah, when you, when you think of that. Speaker A: So, so I like to approach this, you know, the question around AI and crypto in, in, in like two parts and, and two pairs. So pair number one is, you know, you can approach it in terms of what AI can do for crypto and, and also what, you know, what can crypto do for AI, right? And, and then separately, and another pair is, um, just looking at the fundamental properties of each technology. And I think you can contrast the two by saying that AI is a technology of abundance, right?

You can create, you know, anything that, you know, you can possibly imagine, um, with, with AI. And, and the result is just an explosion of new information, new creativity in the world. And crypto, on the other hand, is, is a technology that is fundamentally about scarcity. Speaker B: Yeah, totally. Speaker A: Right. Scarcity, digital assets, the ability to own digital property is built on top of this idea of scarcity. So that's a very high contrast way to pin the two off each other. And I think abundance versus scarcity is kind of why these two things are going to— are such a good fit and are going to propel one another forward.

Let me come back to the first pair, right? What, what AI can do for crypto. One thing crypto is very good at is global, you know, fast, cheap payments, right? We talked about that with stablecoins. Um, I think we're very early in starting to see an explosion of AI agents that can do these very micro, highly specialized tasks. It's my view that those agents are going to want to be paid for their work, and that traditional payment rails historically, you know, have not supported micropayments very well, um, you know, because the fees often, you know, can exceed the value being transferred.

And so as the internet evolves into a network of agents, I think, you know, interacting with each other, you know, via APIs, I do think that stablecoins will end up being kind of the preferred way to settle these very small automated payments, giving agents their own wallets to figure out how to transact without you know, oversight. So that's an example of, of, of what, um, sort of, uh, you know, crypto can do for AI, right? You can enable these kind of, um, micropayments. On the other side of things, um, huge amount of volume in crypto is trading, right?

And, um, a lot of the trading, especially meme coins, for example, um, is driven by sentiment. Yes. Unstructured data that you, you know, you find on Twitter, on Telegram chats, on Reddit. And one thing AI is very, very good at is parsing all this unstructured information in the world and synthesizing it down for you. And so, you know, I, I do think that, um, AI will increasingly play a role in financial markets, including crypto. And we're starting to see more and more teams experiment with that, starting with the thing that's been working— one of the things that's been working in crypto, which is speculation— and using AI to, uh, make investing more efficient, um, and, and drive better outcomes.

So that's, you know, what AI can do for crypto. I've kind of saved the best for last, which is, um, and, and, and this is back to the camp of what, what can crypto do for AI? So, um, needless to say, you know, AI models, foundation models, large language models are very expensive to build. Um, they need a huge amount of upfront cap— upfront capital invested in, in hardware. And energy, right, to train models, to run inference. And today it's the case that if you want to go work on one of the, you know, the frontier models, you need to go work at one of 5 companies basically, right, in, in the West, right?

Google, Amazon, you know, Meta, OpenAI. Crypto offers an alternative to that world. Um, you know, one thing that crypto is very good at is using incentives to bootstrap massive networks of machines, right? Bitcoin is, I think, the most powerful computing network on the planet. And that's because, you know, miners are able to earn, you know, this global asset. I think that what we're going to see is sort of a similar phenomenon play out where we will build one of the biggest computing networks in the world, a network that is specialized in compute for, for AI, um, both for training and for inference, um, where a token is essentially coordinating, you know, the, the running of these machines.

Yes. And the result of that is, is we can build actually OpenAI. We can build networks, uh, sorry, build models that are not controlled by a single company, where any developer anywhere in the world can contribute to them without having to go work at one of those 5 companies. And, you know, this is exactly what Pluralis, which is a portfolio company, is working on. That team is 5 PhDs in machine learning. All of them came out of Amazon's Applied Machine Learning Research Lab. It was previously thought that, you know, doing training of large language models over a decentralized network was impossible.

It was thought that you had to do it kind of concurrently in one data center. These guys have proven that impossibility to be false. And, and so that's, that's probably the most exciting, um, you know, idea of what crypto can do for AI. We can build actually open AI, um, that rivals the, the best proprietary models because you have machines all over the world and talented people all over the world able to contribute to their development. Speaker B: Yeah, totally. Speaker A: Right. Scarcity, digital assets, the ability to own digital property is built on top of this idea of scarcity.

So that's a very high contrast way to pin the two off each other. And I think abundance versus scarcity is kind of why these two things are going to— are such a good fit and are going to propel one another forward. Let me come back to the first pair, right? What, what AI can do for crypto. One thing crypto is very good at is global, you know, fast, cheap payments, right? We talked about that with stablecoins. Um, I think we're very early in starting to see an explosion of AI agents that can do these very micro, highly specialized tasks.

It's my view that those agents are going to want to be paid for their work, and that traditional payment rails historically, you know, have not supported micropayments very well, um, you know, because the fees often, you know, can exceed the value being transferred. And so as the internet evolves into a network of agents, I think, you know, interacting with each other, you know, via APIs, I do think that stablecoins will end up being kind of the preferred way to settle these very small automated payments, giving agents their own wallets to figure out how to transact without you know, oversight.

So that's an example of, of, of what, um, sort of, uh, you know, crypto can do for AI, right? You can enable these kind of, um, micropayments. On the other side of things, um, huge amount of volume in crypto is trading, right? And, um, a lot of the trading, especially meme coins, for example, um, is driven by sentiment. Yes. Unstructured data that you, you know, you find on Twitter, on Telegram chats, on Reddit. And one thing AI is very, very good at is parsing all this unstructured information in the world and synthesizing it down for you.

And so, you know, I, I do think that, um, AI will increasingly play a role in financial markets, including crypto. And we're starting to see more and more teams experiment with that, starting with the thing that's been working— one of the things that's been working in crypto, which is speculation— and using AI to, uh, make investing more efficient, um, and, and drive better outcomes. So that's, you know, what AI can do for crypto. I've kind of saved the best for last, which is, um, and, and, and this is back to the camp of what, what can crypto do for AI?

So, um, needless to say, you know, AI models, foundation models, large language models are very expensive to build. Um, they need a huge amount of upfront cap— upfront capital invested in, in hardware. And energy, right, to train models, to run inference. And today it's the case that if you want to go work on one of the, you know, the frontier models, you need to go work at one of 5 companies basically, right, in, in the West, right? Google, Amazon, you know, Meta, OpenAI. Crypto offers an alternative to that world. Um, you know, one thing that crypto is very good at is using incentives to bootstrap massive networks of machines, right?

Bitcoin is, I think, the most powerful computing network on the planet. And that's because, you know, miners are able to earn, you know, this global asset. I think that what we're going to see is sort of a similar phenomenon play out where we will build one of the biggest computing networks in the world, a network that is specialized in compute for, for AI, um, both for training and for inference, um, where a token is essentially coordinating, you know, the, the running of these machines. Yes. And the result of that is, is we can build actually OpenAI.

We can build networks, uh, sorry, build models that are not controlled by a single company, where any developer anywhere in the world can contribute to them without having to go work at one of those 5 companies. And, you know, this is exactly what Pluralis, which is a portfolio company, is working on. That team is 5 PhDs in machine learning. All of them came out of Amazon's Applied Machine Learning Research Lab. It was previously thought that, you know, doing training of large language models over a decentralized network was impossible. It was thought that you had to do it kind of concurrently in one data center.

These guys have proven that impossibility to be false. And, and so that's, that's probably the most exciting, um, you know, idea of what crypto can do for AI. We can build actually open AI, um, that rivals the, the best proprietary models because you have machines all over the world and talented people all over the world able to contribute to their development. Speaker B: Wow, interesting. I've gotta read more about Pluralis, I think. Um, Okay. I could dig so much more into this, but, uh, for the sake of, of, uh, efficiency, I want to ask about sort of two big narratives of the last cycle and where you think they're sort of at today, what the sort of 2.0 versions of them might look like.

And, and both are things I think you in particular have spent, uh, a lot of time thinking about. And I'm, I'm sort of talking about NFTs and DAOs. Uh, NFTs, I know you're thinking about because you're tweeting about what the, the next NFT marketplace should look like and asking, you know, for suggestions. So I suspect that one's a little top of mind. And then you had a really interesting post about sort of like what the 2.0s of DAOs look like. So maybe we should start with NFTs, you know, with the, the sort of profile picture phase a little bit behind us, where, where are we now?

What are we sort of like, uh, what needs to happen for this to become more useful or more important than it is? Speaker A: Right. So, so again, you know, NFTs are kind of a flavor of the Carlota Perez hype cycle. In 2021, they kind of went through their boom-bust, and, um, today I'd say they're probably somewhere in the trough of disillusionment. They're not yet on the steady slope, um, and productivity like stablecoins and, and DeFi and Bitcoin. The way that I've always thought about the opportunities with the opportunity with NFTs is, um, that NFTs can be the port of entry for all digital media.

Um, so, so not just like collectibles, highly valued speculative collectibles, but, but really like foundational, um, kind of infrastructure that enables true digital ownership and exchange of any type of digital content. Speaker B: And you really worked on this even before Variant, right? Speaker A: Yeah, so 2014 I started a company called MediaChain where we had kind of Same idea, right? To make every piece of media, bring every piece of media in the world on-chain so that you could kind of know where it came from, its provenance, you know, what it was about.

All the information about that media could be accessible to anyone, just like a token is accessible to anyone instantly anywhere in the world. And so, you know, as a primitive, I think NFTs are still incredibly promising. And I think what we saw last cycle is that the kind of speculative fervor around collecting NFTs really squeezed out any thinking around the kind of utility of the primitive itself. I continue to believe that NFTs are going to be a thing. The number of actual non-fungible tokens on blockchains is going to grow many, many orders of magnitude over the next 10 years.

But I think what we've seen is there's less interest in the kind of consumer speculative side of that. And hopefully there's going to be more interest going forward in kind of the developer opportunity. And yeah, I recently tweeted, you know, asking, you know, what are people thinking about building in and around NFTs in 2025? And I think the best response to the tweet was right along these lines, which is like, in 2025, you shouldn't necessarily be building a quote-unquote NFT marketplace. You should be building, you know, a product that has need for some digital, you know, digital asset that benefits from, from ownership.

And, and so, you know, maybe that's a game, right, where all the in-game items are, are NFTs and, and the market around those, you know, makes the gameplay better. So, so, so the concept of like a horizontal NFT marketplace for, for speculation, I think, was a 2021-22 phenomenon. The new opportunity is like, you know, just build a product people want that, that utilizes ownership as a keystone of the user experience. Speaker B: Yes, I got it. It's not going to be the— the user behavior won't be Hey, I'm going to step into this digital gallery or storefront and sort of browse the different pictures or, you know, uh, items that I might want to collect.

And instead it's going to be, I'm using this application, which happens to create, you know, digital property that is given to me in some function. Speaker A: Yeah. I mean, the headline story here I think is, is broader than just NFTs. It's, you know, the shift again, it's, it's from crypto as the, the, the what crypto was this new invention. Everyone was like, oh. What is that? And learned about it. And then the speculative kind of fervor took over to today where crypto is the how, right? Stablecoins, you know, to most users are just dollars.

You don't need to know or care that it's running on crypto rails. Similar thing is going to be the case with NFTs where it's just how it works. You don't need to know or care that it's crypto. One exception that I think is an important one, digital art, you know, didn't have a market before NFTs. I think there is a really strong product market fit around digital art and NFTs today. You know, I have a bunch of digital art on my walls and in our office and a lot of the artists whose work we've got, you know, they didn't have a means to monetize their work prior.

And so that's going to continue to be one area where crypto is both the how but also the what, right? You're, you're buying the digital work and you're buying it because it's an NFT on the blockchain. But for the most part, I think that the expansion and growth in NFTs is not going to be like that. Speaker B: Amazing. Uh, let's talk a little bit about DAOs. What, what were the lessons from the last cycle, and, and how does the, like, 2.0 start to address that? Because there was a lot of excitement, but, uh, I think a lot of them just got lost in sort of designing by committee and, and the, the struggles of that.

Speaker A: So DAO, you know, for those uninitiated, stands for decentralized autonomous organization, DAO. Um, and Vitalik Buterin, the creator of Ethereum, um, you know, I think coined this term, um, in a 2014 blog post wherein, you know, he, he offers a definition, um, that, that I really like that I think we're finally starting to get back to now and that we really got away from. In 2021. And in his definition, he puts a lot of emphasis on the A in DAO— automaton, right? Decentralized autonomous organization. And he says, you know, automation at the center, humans at the edges, right?

And the A is at the center of DAO for that reason. In other words, you know, DAOs should really lean into automation, which is the superpower of smart contract blockchains like Ethereum and Solana. Um, it's also the superpower of AI, right? Automation. And in 2021, we had a lot of experiments around DAOs, um, but I would say almost all of them had very little automation at the center. And in fact, it was humans at the center— was the opposite of what Vitalik was saying— was humans at the center and no automation.

Um, and so, so the 2021 version of DAO was an organization that was kind of owned by users but functioned very much like a classic kind of cooperative, you know, and, and not in the charitable sense, right? Like the best cooperatives, the biggest— there are multi-billion dollar cooperatives— all of them have adopted a management structure that looks like a corporation, which is kind of a tested and true way to scale an organization. The, you know, cooperatives that haven't succeeded typically Um, you know, one reason they fail is, is because they are being managed by committee.

There's a lot of bureaucracy. They're, they're slow moving. And that's, I think, what we saw happen to 2021 DAOs. They're member-owned or community-owned organizations. Yeah. Without any automation, with a lot of humans trying to, you know, manage a thing by committee. And, and ultimately that was slow, not competitive, didn't work. DAOs 2.0, I think, will go back to this, uh, original definition where we will have automation at the center. The automation will be the result of improvements in both, you know, blockchain, smart contract blockchains, and AI. And, you know, I mentioned earlier, I think we're going to see an explosion in the number of agents.

These agents need things from humans, right? They're, they're automating things, but they also need things from humans at the edges that they can't automate themselves. And the opportunity with crypto is to incentivize humans at the edges to contribute. What these agents need. And so in short, I think the opportunity is to finally realize the original definition of DAO, um, you know, what I've been calling DAO 2.0, which is real automation at the center, specifically AI at the center, and AI paying humans at the edges, um, to do work that the automaton itself cannot do.

And, you know, actively looking for, for people working on this, uh, we've got one investment which is kind of a cute example of this called Bato. Which is an AI artist, right? And it's incentivizing humans at the edges to contribute to training its model in terms of what art it should produce and so on. But I think the idea is much, much broader than that. Essentially any economic activity that can be automated and needs human input could be structured as a DAO. And so I think we will realize that idea in the fullness of time, and maybe we're just getting started now that the two technologies are more mature.

Speaker B: And so a DAO becomes in that respect, sort of a pool of human labors for the AI to tap. And the reason it's useful for there to be a DAO structure is that, I don't know, you, you don't want to have to keep finding people each new time you need this help. You have some consistent flow of work that needs humans. Speaker A: First of all, anyone can contribute, right? As you said, it's permissionless. Anyone can contribute. And that's true because the agent is, is using, you know, a global permissionless rail, which is crypto.

Where you can pay anyone, you can kind of, you know, structure your incentives in a transparent, reliable way. And so yeah, that, you know, that's, I think that's always been the opportunity is you can have this automaton kind of incentivizing humans. And by the way, you know, the original DAO, and I think, you know, this fits the definition of DAO 2.0 that I'm offering, is Bitcoin, right? Like Bitcoin is this core protocol where at the center you have this automation, which is solving this very complex hashing algorithm and then paying the humans at the edges who are running that hashing algorithm for you, for the network with Bitcoin incentivizing them to secure the network.

So Bitcoin is the original DAO and it's what we should be getting back to with DAOs 2.0. There's no AI in Bitcoin, but it's, it is fully deterministic and automated at the center. And there's humans at the edges who are making it all work. Speaker B: Amazing. Well, with that, I'd love to get into our sort of, uh, wrap-up questions, which are on the more abstract side. Um, to start, if you had unlimited resources and no operational constraints, what is an experiment you'd like to run? Speaker B: Amazing. Well, with that, I'd love to get into our sort of, uh, wrap-up questions, which are on the more abstract side.

Um, to start, if you had unlimited resources and no operational constraints, what is an experiment you'd like to run? Speaker A: You know, that the— I love the freedom of the question. And, um, so the answer I came up with has actually nothing to do with, with crypto. Um, and, you know, I'd say my— in what I'm obsessed. I've got a green thumb. And if I had unlimited resources and no operational constraints, what I'd love to experiment with is just planting trees at mass scale in urban environments. I think like it would be an amazing experiment to see kind of the effect that that can have.

This was, by the way, in, you know, in part inspired by my own kind of interest, but also Paul Graham had a great post around planting trees in small towns and how I can impact. Um, no kidding. Just like, you know, there was some study done, I think, that, you know, planting trees actually increased the economic productivity of— Speaker B: wow, no way. Speaker A: Um, so totally random, but like, that's what came to mind when I thought about this. It's, you know, if I had unlimited resources and no operational constraints, meaning I had all the time in the world, like, I think that would be a really productive experiment to run at scale, huge scale.

Speaker B: I love that idea. What's the closest you've seen to that in a cityscape? Like, is there any city that you're like, wow, they actually have done a really good job at this? Speaker A: I gotta say, I think, I think New York has, has, has been doing a really good job. Um, they've been planting a lot of trees there. I think there, there's some plan to plant a quote unquote urban forest. So they're like scaling it up now. But, but you know what, I, I would, I, the, the, the no operational constraints and unlimited resources are key here because I, I haven't dug that deep into it.

Perfectly honest, looked into it in a small town where my wife's family is from and astounded to find that, you know, to plant one tree costs $10,000 because like on a size, you know, town side. And so, you know, that, that got me thinking like, there's like, that can't, you know, there's an opportunity here to fix that. Speaker B: I love that idea. What's the closest you've seen to that in a cityscape? Like, is there any city that you're like, wow, they actually have done a really good job at this?

Speaker A: I gotta say, I think, I think New York has, has, has been doing a really good job. Um, they've been planting a lot of trees there. I think there, there's some plan to plant a quote unquote urban forest. So they're like scaling it up now. But, but you know what, I, I would, I, the, the, the no operational constraints and unlimited resources are key here because I, I haven't dug that deep into it. Perfectly honest, looked into it in a small town where my wife's family is from and astounded to find that, you know, to plant one tree costs $10,000 because like on a size, you know, town side.

And so, you know, that, that got me thinking like, there's like, that can't, you know, there's an opportunity here to fix that. Speaker B: Yeah, totally amazing. Which tradition or practice from another culture or era do you think we should bring back or adopt more widely? Speaker A: In America, I think there's this little-known tradition from way back around the turn of the century, Industrial Revolution, around community ownership or cooperative business structures. You know, so, so while like the West was getting settled and industrialized, right, small towns, for example, like in, in often got their power, their electricity by virtue of like setting up a cooperative utility.

Company because, you know, that was kind of the only way to fund and manage a utility at small scale. Same thing with agriculture, right? You know, farmers kind of pooled resources to create kind of a community-owned dairy cooperative or something. And, you know, today Land of Lakes Butter is like a multi, multi-billion dollar cooperative. Some of these things have really scaled up. That's a tradition that's, you know, I think less studied in America, America being kind of the most capitalist country when you, when you hear cooperative. Very often people immediately jump to, oh, that's kind of socialist, or— but in fact, like, there's, there's a rich tradition of a cooperative tradition in America, and it very often kind of coincided with these kind of small, you know, communities that were looking to, you know, leverage markets to actually get their product out.

Like, I think there's a lot to learn from that, and I do think that crypto has the opportunity to scale it up, scale up this idea of community ownership to internet scale, particularly if we leverage automation and we leverage free markets and incentives to enable global participation. So crypto, in short, combines the best of free market capitalism with this kind of cooperative tradition of community ownership. And I think the result is we can use market incentives to build networks bigger, faster, than ever go forward. Speaker B: Fascinating. Okay, final question.

If you had the power to assign a book to everyone on Earth to read and understand, what book would you assign? Speaker A: It's, it's Narrative Economics by Robert Shiller, and it just explains, um, by way of a number of examples, um, how narratives can drive economic outcomes that really change the world. And it, you know, he gives the example, one of the examples is a crypto example, which is Bitcoin, and, and it sort of gives the example, all, all the components Bitcoin narrative that have led to its success. And I would say it's, it's, it's Bitcoin's success that has opened up the possibility for all these other networks that, and you know, things like Ethereum and, and, and others have followed down that path.

And that in turn has enabled all this, you know, programmable permissionless innovation on top, like stablecoins and so on. So I think that's a good foundational read for anyone looking to understand the genesis of the crypto space and, and sort of what could potentially still come. Speaker B: I love that. Um, I thought non-zero chance you pick Satoshi Whitepaper. Uh, and so this is, you know, uh, an extra layer that I, that I love that, that frames it all, uh, in a bigger, bigger canvas. Speaker A: The, the debate, honestly, you know, I had to pick one, so I picked, I picked this one, but the, the, the runner-up was the Ethereum Whitepaper.

Hmm. Speaker B: Okay. Speaker A: I'd highly recommend that too. The Ethereum Whitepaper is, uh, I think it reads like a sci-fi novel. Um, it's, it's one of the things that got me so excited about the space. I think a lot of the ideas there are going to come to fruition in the fullness of time. So definitely recommend that as well. Speaker B: Okay. Speaker A: I'd highly recommend that too. The Ethereum Whitepaper is, uh, I think it reads like a sci-fi novel. Um, it's, it's one of the things that got me so excited about the space.

I think a lot of the ideas there are going to come to fruition in the fullness of time. So definitely recommend that as well. Speaker B: Amazing. Well, Jesse, thank you so much for spending your time with me and, uh, and talking through so much of this. I really appreciate it and I really enjoyed it. Speaker A: Yeah, likewise. Great questions. Thanks for having me. That's it. Speaker C: Thank you for listening to this episode of The Generalist Podcast. Please subscribe. On Apple Podcasts, Spotify, or your preferred podcast app. Ratings and reviews help others discover these discussions, so if you enjoyed the conversation, I'd be grateful if you could take a moment to leave one.

For all past episodes and more, visit us at com. See you next time as we continue to explore the future.