Solana 的 Anatoly Yakovenko 谈加密货币的下一个时代：量子、AI 与货币的未来

Anatoli is the CEO of a little crypto project known as Solana, one of the fastest growing blockchains in the world. As CEO of Solana Labs, he's driving web 3 innovation. BlackRock, the world's largest asset manager, expanded its $1.7 billion tokenized money market fund to Solana. Why don't we all switch to Solana? I mean, Solana sounds like it's actually commercial and the other guys sound like they're antique. Everybody in the world should be your customer. Crypto will eventually win. It's inevitable.

Ladies and gentlemen, please welcome Solana co-founder Anatoly Yakovenko.

Oh man, thanks for having me.

How are you doing? Thank you. Welcome.

Thank you.

How much of a difference has David Sacks made in the first 6 months as crypto czar for your industry?

Oh, it's been incredible. I think it's night and day. I don't know if the industry would have survived another 4 years of the Gensler regime. The GENIUS Act, I think, is going to unlock an estimated $1 trillion to $10 trillion worth of stablecoins that are going to be on public, permissionless chains.

If you look at the charts of who owns Treasuries, it's China, Japan, and other countries right now. I think Tether is somewhere around number 5. Within 5 years, I think the internet is going to be the largest holder of U.S. Treasuries. At such a scale, I think—I'm an engineer—I cannot honestly comprehend how that's going to change finance, but I think it'll be transformative.

What's the upside and downside? Are there concerns there as well with that huge impact on democratization, or are you kind of a libertarian—let the chips fall where they may, so to speak?

I think it's a huge opportunity to really accelerate American innovation and spread American finance around the world. I think we actually have the best financial system in the world. It's the most trusted, the most robust, and the best regulatory environment, for what it's worth, as well.

But it was built after World War II, before the internet, so its APIs are kind of fax-machine-based. What crypto is allowing, I think, is this new technology stack built on top of the internet that's completely Western-aligned. It's for transparency and capitalism, but now we can actually interface Western, U.S.-based finance with the rest of the world. I think America is going to benefit primarily from this.

Similar to our media business going around the world and infecting people's consciousness.

When you were getting Solana off the ground, how much of it was a technical and architectural vision that you had, versus maybe a set of trade-offs that you were trying to solve that Ethereum didn't fill or Bitcoin didn't fill, and you said, “I'm just going to try and do this”?

I can't speak for all founders, but I think founders are driven by a crazy vision. They have to be a little bit insane. My insane vision is always this idea: Imagine finance 20 to 50 years from now—the science-fiction version of finance.

What I imagine is a single, giant ledger, a single computer for every market in the world. That means it's available in Nairobi, New York, London, and Singapore, and all of these things are synchronized at the round-trip time of the speed of light through fiber around the world or through Elon's satellites. That's 120 milliseconds. A dollar can be in New York, London, Singapore, or Nairobi in 120 milliseconds, so the velocity of money and the velocity of assets are as fast as physics allows.

This is what nerd-pilled me on building this. It's a physics problem. It's a massive finance problem. It's a really fun, low-latency engineering problem.

Did you feel that you had missed it somehow?

When I had my Eureka moment and did the back-of-the-envelope calculation for the design, I thought, “Oh, this is 1,000 times faster than ETH.” When I started talking to folks in the ETH community, they were focused on settlement. Settlement doesn't have these latency problems. You can do settlement in minutes, and that's fine.

I always felt that Ethereum was the world's settlement layer, while Solana is the world's execution layer.

Yeah, so far, so good. Execution is where all the money's made.

So, I think we're on the right track. A fast execution engine can also do settlement. That's kind of a feature.

You've been super critical about 2 things: memecoins, even though they do throw off some revenue for Solana, and also the idea of a crypto strategic reserve. What about those 2 things tweaks you a bit?

I think, primarily, we could not predict what was going to happen on-chain. We called it “blockchain at Nasdaq speed.” That was our tagline, and the idea was always: How do we get stocks, bonds, Treasuries, and real-world assets on-chain from all around the world, to be traded by everybody around the world?

But it turns out that is a much harder legal and regulatory problem than it is an engineering problem. Anybody in the world can create markets for anything, including memecoins and NFTs. Those things took off, I think, in part because regulation was slow to catch up.

Which makes it annoying that those are the things that come out instead of your true mission.

Yep.

We saw Adena from Nasdaq here yesterday. She announced the tokenization of securities that we're going to trade on the exchange. There seem to be a lot of regulated exchanges and businesses from deeply regulated backgrounds starting to experiment with blockchain technology.

Do you think they're going to be advantaged or disadvantaged, given where they're coming from? Does the lock-in and relationship with regulators, along with the lock-in with market participants, give them some leg up? Or do you think the disruptors are ultimately going to be able to operate more freely and more quickly?

This is the big challenge. I think the advantage that we have is that we're very nimble and can operate everywhere in the world. The advantage they have is that they're already regulated. They're already operating with the assets that we want on-chain in the United States, but they don't have global availability.

Nasdaq is still in its little sandbox, so we'll see what happens. I think once the regulators allow public-key cryptography to manage and transfer assets, that's the interface that you can wrap around and start moving anything from inside Nasdaq to Solana and vice versa. Once that interface exists, I think the genie's out of the bottle. The toothpaste is out of the tube.

Do you ever meet with the regulated exchanges, and are there ways to build integration and partnerships that benefit both?

Of course, yeah. We've talked to folks across the spectrum, from banks to regulated exchanges and regulators themselves.

Solana is fundamentally a protocol. It's like an email standard. It's a bunch of software. The people who run it don't report to me. I can't fire them, so I can't stop it if I wanted to.

If we succeed—if the protocol is awesome, globally synchronous, and super fast—Nasdaq would make more money by just running a Solana node and integrating with it more directly. To me, it's ultimately a win-win. We're never going to build an exchange that is onboarding U.S. institutions and serving U.S. customers. We want Nasdaq to do that, run it on Solana, and that would be great.

There's a common claim by the masses—meaning not everybody that's all-in on crypto—that it's still extremely complicated to understand. Even if it's just minting and burning or yield farming, you say it to a normie person and their eyes glaze over.

What's the turn in the abstraction of all of this stuff that makes crypto truly mass-market?

I actually think that the human brain has to change to adjust to it. I agree with you: It's really complicated. But I landed in the States in 1992 from the USSR, effectively, and there's no way my parents could understand what a web link was.

Whenever you have a new technology, it just takes people a long time to adopt it and build a mental model for it. But now they do. They understand the web after years of using it.

As stablecoins proliferate into the back office of a lot of companies, people will figure out, “Oh, this secret key is actually really important. I need hardware. I need PKI. I need trusted displays.” All of the security stuff. They will build a mental model for cryptography and having true ownership over something that is globally transferable.

I saw a chart recently that showed that the number of L1 and L2 projects keeps growing year over year, over the last 3 or 4 years. Why is that happening?

What need are they filling?

Well, I think the opportunity is so big to be the Google of finance, right? If you're the one place where all of finance and all markets run, that is a massive opportunity.

People are going to keep launching L1s and L2s. They're all competing with Solana, and that's fine. I love competition until somebody wins it. As long as we're laser-focused on improving the product—making it faster, cheaper, and more reliable—we have a really good shot at actually becoming that global execution engine that's serving all of finance.

Outside of finance, what do you think is the vertical that has the most promise over the next 5 years?

In crypto or in general?

In crypto, whether it's for Solana or any crypto project, where do you think people aren't putting enough attention?

I think all the stuff that people have tried is kind of like the early days of experimentation. Friendster—all of those things failed until there was a critical mass of people who understood how the web works, and then Facebook took off.

So I think even the weird experiments with NFTs being a way to create a community of artists to build a movie or story and create truly new IP, all that will happen just 5 or 10 years from now, once we hit critical mass.

So, a lot of false starts, and somewhere in that graveyard you might find some really good ideas.

Absolutely.

Just like what happened, the social network concept always seemed to me to be such a winner. Whether it would be like a Digg or a Reddit format where you could vote things up with a cryptocurrency, your comments were somehow related to that. There were a couple of little experiments I remember looking at for investment, but candidly, I didn't think the founders would pull it off, and I was right in that case. Is that the one that you think could break out if Elon put Dogecoin into X, or put in Solana, and there was some sort of currency inherent to the objects and the behaviors?

I personally think that you could build a competitive product to TikTok with crypto if you can catch that kind of lightning in a bottle, because the monetization mechanism with crypto is so different from the ad-based one. The ad-based one kind of creates this forcing function for a lot of spam and duplication to rise to the top.

How would that work? Just describe your product thinking there—that new kind of experience. How do you think it would work?

I think you're seeing some of these things play out with memecoins, where you have creators who are associated with a coin that continues to have market cap and traction. Now, the regulatory environment isn't here yet to clearly tie the success of that creator to the value of that coin. You need to remove a whole bunch of bottlenecks there, but the product exists. People watch that particular creator stream and go buy that coin. Once it actually looks like an investment thing that Jason would be like, “Okay, I have all legal protection to actually put money in here,” I think—

This is related to the financing question I was asking Neil and Ari about, which is: Can creators raise funding this way, and then can they deploy that funding, but the coin holders can actually have equity in that project and in the performance of that project over time, rather than it just be—

If the regulatory environment changes, there's this project that I love, Clanosaur. These are cute little dinosaurs that kids love. It looks like a Pixar dinosaur. They've won awards for their animations, and they raised funding because they created this collective set of dinos. Now, it would be awesome if those dinos could actually have copyright and revenue association in the future, but we can't do that yet, and that's frustrating. It could totally happen once we have enough clarity.

Well, just imagine: We bought collectibles, and if we all bought Marvel comics when we were younger, but we had equity in Marvel, 30 or 40 years later those characters hit and you own it. It could be MrBeast, or it could also be the next creator. You're watching an up-and-coming creator. You want to bet on that?

Is that the next piece David Sacks is working on?

The Clarity Act is the big piece.

Explain it for everybody.

So, again, I'm an engineer. From my lens, it's about raising money in the US and trying to launch a token. We raised a seed round. It was about $14 million, which is amazing. It was outright crazy success for a new, first-time founder. I had to spend $2 million of that on lawyer fees, which is more than 10% of my runway, to figure out how to launch a token in the United States.

Because I have kids in the US, this is my home; I'm never leaving it. So I had to do it in America. A lot of founders actually just left to do it outside of the US. The Clarity Act is a whole bunch of complicated legislation to try to minimize, hopefully, that cost and make it much easier for founders to launch.

It's far too much friction right now.

Provides clarity.

Our partner David Sacks launched a company a few years ago that was trying to tie crypto to real estate as a real-world asset. Tell us about what that movement is all about and what utility is there if it works.

People want real-world assets on-chain because there's demand in DeFi for non-correlated assets. If everything in crypto is a memecoin, all this innovation around real-time risk management between borrowers and lenders is useless. Everything's correlated, so it'll all crash at the same time. There's no hedging, right? The only free lunch in finance is uncorrelated assets if you have true hedging.

So we need real estate, bonds, insurance, whatever have you that has—

Oil. Exactly. Commodities.

But even California fire insurance. It would be awesome to put that on-chain, because then people could actually buy insurance. All those assets, if they exist in this kind of global, synchronized, giant state-machine environment, can all be used together to reduce risk for the entire system because they're uncorrelated. That's actually the only free lunch you can get in finance.

There's a lot of demand for them, and the technology is there to leverage them. Now we just need the regulatory side to catch up.

Can I change tracks a little bit? You're an engineer. You work in cryptography. Have you looked into quantum projects? What do you think is the state of development in quantum computing? Everyone's got a different story. How much is hype and marketing, how much is real, and what do you think is going to happen over what period of time?

Honest answer: I feel like there's a 50/50 chance that within 5 years there is a quantum breakthrough. Part of that is because of how fast AI is accelerating.

Define breakthrough.

You can run Shor's algorithm.

Yeah, we should migrate Bitcoin to a quantum-resistant signature scheme. This is my bet, and this is because there are just so many technologies converging right now. The asymptotic rate of AI, and how fast it's accelerating from a research paper to an implementation, is astounding. I would try to encourage folks to speed things up.

My key for this is Google and Apple adopting a quantum-resistant cryptographic stack. This is the time to migrate, because now the consumer side of it is effectively solved and you don't have to kind of—

So you watch where Google’s going.

Yeah. But I would—I think if you're in the field, you should be worried. For the general public, quantum computing is such a massive unlock in terms of how much we can process that it's going to be as big a wealth creator as AI if we pull it off.

So I think this is, to me, a lot of engineering work. We have the right people to do it, but for everyone else, it should be a huge opportunity.

But to your point, the reports on the breakthroughs on the Willow project at Google are driven by AI modeling, and AI is unlocking a lot of the capabilities to make it real, which seems to be an accelerator. It's pretty powerful.

What's the intersection of all of that world—AI in general and crypto?

This is a funny thing to ask because I feel like AI is going to be everywhere and crypto is going to be everywhere, but where those lines cross is really, really hard to pinpoint. I don't want to say something lame like, “Oh, we have agents sending money around,” because that's kind of obvious.

I mean, I think the first attempt was to say maybe there are distributed networks of compute, and maybe we can run distributed learning or distributed inference, but those projects really haven't—

Taken off and really generated any momentum.

Yeah, not yet.

And again, because they're competing with a data center that is all colocated and funded with traditional finance. You can put those assets on-chain, and that's a lot of ways how I think things are going to integrate.

Probably the most singularity bet we can make is to have an agent that is a creator, that is an X personality, that you can interface with tokens and buy into and pay for the GPUs. That could be fun. But—

Bitcoin has turned out to be surprisingly resilient, but now we're starting to see certain players corner the market on large percentages of it. That was never supposed to happen. If something like MicroStrategy owns 6%, that's actually maybe 50% more than that because there are so many dead coins out there. Does that worry you, the centralization of Bitcoin? Does that mean there's an opportunity to start the game anew?

I think Bitcoin is resilient to these entities collapsing. It's not going to be without painful risk in terms of the people who own Bitcoin, but the thing is, it'll survive that, and all the properties of Bitcoin that people value will remain through that transition. So if you really value Bitcoin, you should see that as an opportunity to own more of it.

Even if somebody were to own 20% or 30%? It seems like there are people who actually have this intent. That's why I'm asking.

Yeah. I think as long as it's an open, global competition to acquire Bitcoin and anyone can participate in that, and we don't end up in some kind of regulated nightmare—you know, like you can't acquire gold or something, like in the ’70s—I think Bitcoin would survive those kinds of shocks.

Is Bitcoin valuable enough now that it makes sense for—I guess North Korea does this, but I was just going to generalize and say—state-sponsored ways of either trying to penetrate it, hack it, or take individual accounts? It just seems like there's an emergent trend here.

Its beauty is that it’s the simplest protocol you can build because it’s focused on just settlement. It’s very easy to understand from an engineering point of view, and proof of work is brilliant. It’s a masterpiece in terms of elegance and simplicity, and it’s very robust to all sorts of attacks.

Now, that doesn’t mean that you can’t have an attack that could cause an unexpected rollback, but I think it’s extremely hard to pull off. It’s very unlikely, and the internet is so super-connected that it can automatically respond and take action.

I actually meant more about states targeting accounts that have large Bitcoin holdings, trying to figure out who owns them, and then basically getting them to give them the coins.

Yeah, those kinds of state-sponsored wrench attacks. I think what we should do, living in the West, is really have strong opinions about property rights and how important they are and how foundational they are to wealth creation in the West and America.

Completely agree with this.

And this is our best defense.

I completely agree with this, and we should be hyper-transparent about who owns the coins, because then it’s like you can’t take away something that everybody knows you own. But when you try to hide your ownership of it, it makes it easier for somebody to take it away.

I think privacy is a right, so it’s somebody’s right to be able to do that. But I think our best bet in wealth creation is actually defending these rights and defending the right of somebody to own Bitcoin if they want to.

It’s extraordinary that it hasn’t been hacked with so much at stake. Maybe you could speak to it as an architect yourself.

It is. The reason it hasn’t been hacked is because it’s so simple, and as an engineer, you always strive for simplicity to achieve a certain outcome. You can’t always achieve that. Solana is much more complicated because the outcome we’re striving for is hyperperformance, and it’s just hard.

So Solana is much more complicated as a result of that. But Bitcoin is designed for a very simple settlement layer, and I think the coolest thing—the coolest piece of software written in the last 20 years—is, I would say, the Satoshi Nakamoto implementation.

There’s been an enormous renaissance in the ETH market recently. Where do you think that comes from? Is that market-driven and speculatively driven, or do you think that there’s a fundamental reimagining of where ETH lives, between Bitcoin over here and Solana over here?

Honestly, I’m a huge fan of Ethereum. I think Vitalik is an amazing person, an amazing engineer, and has a very strong vision. It’s very different from my vision for Solana, and it’s really cool to see those 2 play out. If I could predict what I do that could cause a price change, I’d be a lot more successful.

Well, you’ve been pretty successful, too. But it’s just really, really hard to attribute the work that you do.

Okay, so look, your transaction network is quite liquid. It’s going to become more and more and more so as you have more validators, more clients, and all that stuff. Another market that seems to have built a monopoly or a duopoly around transactions, and that’s a little bit at risk, is Visa and Mastercard. What do you think about that?

My contrary opinion is that I think Visa and Mastercard are more technology companies, and if you look at their profit margin on the gross payment volume, it’s like 10 basis points. It’s like vapor. I think the issuer and acquirer banks are the most disruptible pieces in there because their profit margins are like 2%, much, much bigger, right?

And Visa is a technology company that owns the customer end to end. If they could remove the banks from the loop and just do stablecoin transfers behind the scenes, I think they’d become a lot more successful, and they could do a lot more for a lot less.

Long stablecoin, short banks.

I’m not an investor, but maybe Chamath seems like a good premise.

I can’t comment on this.

So that’s a yes. Everybody short the banks. Everybody’s telling you to do this. This is financial advice—unless it doesn’t work out.

Thank you so much. Appreciate it. Appreciate it.

Thank you so much. You’re awesome, dude. Appreciate it.