整个金融业都在上链：Dragonfly 普通合伙人 Rob Hadick | EP 165

And now it all comes down to financial rails. That’s actually the whole story for today. Interestingly, for me personally, this is what worries me. I’ve lived in New York for 15 years and have dedicated my entire career to fintech, payments, and things like that.

Many years ago, I thought, “Okay, the reason I got into this field is for better financial rails.” So seeing it all come down to this, I think, has brought many to a moment of reflection where they’re trying to figure out what they’re doing here. Whereas for me, it’s something I’ve always done, and I’m glad to see it really starting to grow.

When I think about what happens in business, the end user usually doesn’t care, right? The retail user—and the business definitely doesn’t care, right? So if you look at successful projects, they’re the ones that optimize the product and go-to-market, not the ones that pursue decentralization or another ideology. It’s more of a redistribution than just a devaluation of everything.

Yes, that’s generally my view, and I think you see the pressure on the commissions. We are the new middlemen, in a sense, aren’t we? We democratize intermediation because we can share it, perhaps with the young consumer.

That’s what you see from the neobank side. All these neobanks say, “We’ll give you 3%, 4%, 5% cashback.” But how do they subsidize it? They subsidize it through tokenized revenue and interchange. Instead of the issuing bank keeping it for itself, it shares it with its customers.

Rob, thanks for coming to the podcast. We just went to the Out East Summit, in Hampton—or East Hampton. I’m not very good at navigating.

North Fork.

1. Why Speculative Narratives Are Fading

Okay, North Fork. I don’t know this area of New York very well, but the conference was good. I sensed that many people in traditional finance were generally optimistic about stablecoins and trading real assets, but I’m curious what your main takeaways were.

Yes. First of all, thanks for the invite, and I’m glad you’re getting back to podcasting.

Thank you.

The Audi Summit, in particular, is an event very focused on traditional finance. We had heads of digital assets from Morgan Stanley, Invesco, Wellington, and other places, so, quite predictably, the conversation was around, “Okay, what do banks do? What do asset managers do?” I think Franklin Templeton sent about 6 people there.

It’s understandable that people want to talk about what excites them, and things like tokenization are now on the radar of every bank and asset manager. Things like stablecoin payments and using stablecoins as a form of collateral are on everyone’s minds right now, and people are excited.

It’s quite interesting to be in a group of people at a time when, say, you can go on Twitter or talk to more crypto-oriented people and they’re just in a state of desperation. So these are my conclusions: Finance is moving to blockchain, whether or not the rest of the crypto community cares.

Yeah, I feel like this is a great start, because, at least in the past, it was like crypto—there was Web2, then there was Web3, and Web3 was supposed to kind of eat all of Web2. We had the NFT wave, the metaverse, and DeFi summer.

Now it seems that, both from an investment-universe perspective and from a product and revenue perspective, some of the euphoria has died down, and now it’s largely just finance.

Yes, and that’s not bad at all. I think, from my perspective, you can still get results 100 or 1,000 times higher even if we’re just doing finance. But it really seems like the crypto-native community is in despair, while, according to you, there’s a lot of enthusiasm on the side of capital allocators because new mechanisms are emerging that make everything much more efficient.

Yeah, it’s weird, I think, because there’s also this San Francisco and New York factor. A lot of people in crypto have gone through multiple waves. There were old-school OGs who were largely ideological supporters of this. They were perhaps libertarians. It was something like a religion with Bitcoin.

Then, I think, the center of gravity started to shift. In those libertarian times in New York, there were a lot of financiers who started talking about the devaluation of money and really understood this story after the crisis. Then, when Ethereum came along, we started saying, “Okay, what can smart contracts do? What is the future of technology? How can you own part of your identity or part of all these other things that we believed in?”

The center of gravity shifted to San Francisco, and it was about a lot of these larger stories that we were talking about—reimagining the way we did everything. The book Read Write Own was kind of a manifesto that so many people believed in at the time, and 2021 was really built on that.

All these different NFT projects and social projects seemed to take up most of the information space, even as the world of DeFi and so on was also developing. Of course, before that we had DeFi summer. The last 4 years have been very difficult for this community and the people who were focused on this part of the crypto world, but what kept running were the financial rails.

If I think about the biggest companies in this space, they’re the stablecoin companies. If I think about the largest private companies—not just Circle or Tether, but also service providers—again, stablecoin companies or trading companies are the ones that are very successful right now.

Then we start talking about service providers, security companies, and people who help all of these entities operate in the regulated field. Now it all comes down to financial matters, and that’s the main story today.

It’s interesting because, for me personally, these are things that concern me. I’ve been in New York for 15 years and spent my entire career doing fintech, payments, and things like that. Many years ago, I thought, “Okay, the reason I really got into this field was because these are better financial rails.”

So seeing this field come down to this, I think, has really forced a lot of others into a moment of truth where they’re trying to figure out what they’re doing here. Whereas for me, this is what I’ve always been doing, and I’m glad to see that it’s starting to really grow.

2. Finance, Payments & Tokenization as the Only Scaling Verticals

What is his famous phrase? Exchanges, stablecoins, and L1s.

L1s.

Yes, which ultimately comes down to execution.

Yes, 100%. And so I think he’s right. The thing that we also generally looked at is, okay, if it’s trading in general, then things like order flow and where the deals are coming from are going to be important.

I completely agree. I was talking to Thrun the other day, and we were discussing how we were exploring the perfect L1 maze, or just scaling solutions in general. We had Ethereum, of course—the first smart contracts, with relatively low throughput. Then we had upchains, sharding, L2, and L3, and we explored that whole spectrum.

And now I feel like we’re doing the same thing with the trading infrastructure. It’s like, okay, L2s such as Base, whether it’s Arbitrum or the Robinhood Chain, and then you have Hyperliquid, where everything is kind of quasi-hosted, like AWS in Tokyo.

Then you still have more general blockchains with high throughput. Now the question is, “Okay, what’s the best trading or maybe stablecoin infrastructure?” It’s definitely narrowed down to the financial aspect.

100%. A lot of the discussion right now is driven by what’s happening in Washington, and what’s happening in Washington is largely driven by what’s on Wall Street right now and what’s on the minds of big companies.

Even tech companies—they don’t talk about it that publicly, but, for example, Meta says, “Oh, we’re going to implement stablecoin payments.” That’s what they’re talking about now. They’re no longer talking about rebranding to Meta, and they’re no longer mentioning the metaverse, right?

I remember, I think it was in 2019, or even earlier with Libra, there was a lot of anxiety that Meta—or Facebook, as it was called then—would come in, crowd everyone out, and destroy the blockchain industry. Of course, the project never launched, but there was a lot of hostility in the crypto community because it wasn’t decentralized and some big centralized company would come in and take over everything.

Now we’re like, “Oh, we’re glad Tempo is here,” or, “Oh, USD, we’re so glad it’s here.” That’s exactly what people were trying to do not so long ago.

I even remember that in 2014, I was working at Goldman, and we created the first working group on crypto assets and digital assets. It’s funny because it seems so far back, but the conversations were pretty much the same as today. They discussed, “Well, should we support Bitcoin trading?” And also, “We believe tokenization will become important.”

Goldman actually invested in a project called Axoni, which was kind of the first high-performance platform for tokenization—a kind of banking consortium. Most of those conversations are the same conversations I’m having today.

It’s just that we’ve finally gotten to this point. The infrastructure is finally ready, and we’re starting to see real work being done there. But it’s really interesting to me that I’ve spent time in this field, been in it partly while doing other things, and watched the ups and downs while the topics we talk about stayed the same.

Ultimately, I feel like we are exactly where we were thinking and talking about 10 years ago today.

It’s kind of, I think, more like a DeFi mallet that we talked about before—maybe interfaces with KYC if you’re bringing in more shares or different institutional pools, and on the backend, cryptography.

That's interesting. We've been talking about these things for years. We explored different trade-offs, and I think the projects that have generated, say, over $100 million in revenue are mostly financial or trading infrastructure. 100%.

In the off-chain sector, it's all exchanges, stablecoin companies, and some software vendors. If you think about Fireblocks, they're a software vendor, and it's very interesting.

Regarding your comment about the “DeFi mullet,” one of our portfolio support team members asked me yesterday at lunch, “Hey, I hear this from all of our portfolio companies. What do fintech companies really want?” I said, “Who’s asking you that?” These are all DeFi companies. Each of our protocols says, “How do we sell to fintech?”

This is quite understandable, given the events of the last 9 months. Trading volumes have fallen significantly in crypto-native assets, especially in altcoins, except memecoins, and RWA volumes on the network have increased thanks to Trade XYZ, Hyperliquid, Lighter, and others.

3. Institutional Adoption & What Wall Street Actually Wants

Every exchange has become an “exchange for everything” because they support everything. With Morpho and Athena starting to implement storage on Robinhood and Coinbase, this is becoming evident. Now it is seen as the next wave of distribution.

“DeFi mullet” is the future we're working on, and fintech will just become these non-custodial neobanks. The story with non-custodial neobanks has been a big part of the growth of Rain, which is now one of the largest companies in the space, providing cards and wallets for all of these players.

From a revenue perspective, this story has probably been the biggest growth driver over the last 12–18 months across the entire industry.

I want to talk about Rain, but you mentioned fintech, and I'm curious. If you look at traditional finance, a lot of it has been pretty slow to adopt technology. Then the question would be, “Are you going to compete with Chase, Wells Fargo, or another big traditional financial player?”

Even with the likes of Robinhood, it turned out that the real fintech players actually developed these new technologies and became the drivers of innovation. Do you think it will be the same in the cryptocurrency space, or will we find ourselves in a “DeFi mullet” situation, so to speak?

Perhaps fintech companies that already have distribution channels will simply connect to the blockchain infrastructure, and they will gain a larger market share than crypto-native projects.

I believe there will be successful crypto-native players, but there is always the “incumbent’s dilemma” that everyone talks about. Obviously, there are companies like Robinhood, Ramp, or Stripe that are in many ways incumbents, even though they try to pretend they're not. They already have this distribution mechanism built in.

They are rethinking many aspects of their businesses with an eye toward the future of cryptocurrencies, stablecoins, and AI. These companies will be successful. I get asked all the time by people in stablecoin companies, “Why doesn’t Stripe take over the entire market?”

The reality is that Stripe is a big incumbent right now, and you definitely don't want to bet against them. They will continue to be very successful, but first of all, the market is too big. They can't take everything.

Second, we hear about a lot of coordination issues in the Stripe ecosystem because it's still a big company. It's not necessarily about them specifically, but when you reach a certain scale, no matter how much you talk about new directions or rethinking the business, you always have a huge burden hanging around your neck: what you're already doing, all the cash flows that are required, and all the employees and stakeholders who are watching your every move.

There will be established players, and there will be challengers who will come and achieve incredible success, just like in any market. Many of them started out being more crypto-native.

If you look at the neobank sector, there are a lot of startups like Arc in Latin America, the former Dollar App, and companies people know about, such as Cast, Read.pay, E, Erifi, and all the others. Not all of them will survive, but some of them will definitely succeed.

They're all growing very quickly now. Will they replace Revolut? Probably not. But will they compete for a certain market segment? Absolutely.

I don't know what Peter Thiel is saying there. “Competition is for losers,” right? This will quite obviously happen, but it will not be a purely crypto topic.

It's quite funny because I think it's a bit off topic. But in the 21st century, everything—and for a long time, everything, including China—just moved up and down together. It was so correlated that people still expect it to continue, for example, in the venture capital business, but there is no reason for it to.

In fact, everything should be the other way around, so there will be winners and losers. Established players cannot capture the entire market.

Speaking of the infrastructure side and a return, perhaps, to the early ethos of crypto, I posted a tweet—I don't remember, a month or 2 ago. It said, “I don't care about decentralization. I'm concerned about better products.” Then Al Ali from A6Z and a bunch of other people commented, saying that we had gone too far and that it goes against the ethos of blockchain.

For me, the point wasn't that decentralization is bad. I think the industry has perhaps become too focused on decentralization for the sake of decentralization, instead of building better products—even with the arrival of fintech companies like Coinbase with Base, Robinhood with its new L2, or even Stripe and Tempo.

Do you think they will continue to take market share away from so-called “crypto enthusiasts,” perhaps because they are less idealistic about infrastructure? Or do you think some of the ideals of the crypto community still matter?

I'm not sure I ever thought they mattered, to be honest, from an investor's perspective. There is a reason for their existence. These ideals are good in themselves in many ways; they serve a purpose. Decentralization and the ability to coordinate trust are certainly important, both from a censorship-resistance perspective and simply from a freedom perspective.

I don't want to say that they're worthless. But when I think about what happens in business, the end user usually doesn't care. The retail user and the business user definitely don't care.

4. Token vs Equity Value Accrual

When you look at what was successful over time, it was those who optimized the product and go-to-market, not those who optimized decentralization or some other ideology.

There's a well-known way of thinking that a lot of Bitcoin's success is because it has become a religion. Many things are religious, in a sense. People are fanatically supportive of many different things. Even the cohort of value investors, like Warren Buffett, is somewhat religious, even though value investing has performed much worse than growth investing.

Moving forward, we see revenues and users rallying around those who didn't really care about decentralization. I expect this to continue to be true for any significant period of time and for any significant revenue growth.

If you want to win, the best companies are dictatorships. The best companies are not democracies. We talk about corporate governance, and that's necessary, but almost always the best companies are very close to dictatorships.

I've long said that I believe in benevolent dictators. This sounds a bit harsh, but the point is that it's hard to get people to row in the same direction. If someone says, “Hey, we're rowing north, so either take a paddle and row north or get out,” they may be wrong about moving north, but at least they're all acting in concert.

I think that's the hardest part, and that's one of the main reasons I posted that tweet. You see how the industry has largely resorted to various contortions to fit into the decentralization model and, in my opinion, has given up on better products and profits.

It's like righting the ship to focus on better products and profits, even if the original “believers” or “priests” are no longer very happy with it.

Yes. Just look at some of the people you mentioned who responded to your tweet, and some of the investors. If you look at how they've changed their approach to investing over the last 6–12 months, some of it isn't public yet, but we're seeing it because we're following all the rounds.

I can say that the funds, which have now become larger and were perhaps more ideological before, have now become less ideological. They all said, “Okay, we're going to invest in companies that are much more centralized, business-oriented, and dictatorial, so to speak, and that barely operate on blockchain rails.”

You mentioned Ali. They invested in Canton, right? You can argue about how important centralization is and what exactly makes a difference to your business. But of course, there are changes in investors' approach to the market. They're starting to address issues that are already reflected by public investors, token investors, and prices.

Moving on to the product aspect, you mentioned Rain as one of the breakthrough apps, and I know you are an investor in it. What are your thoughts on the product roadmap, given that we're now more focused on product and revenue, and that it's mostly finance, trading, stablecoins, and execution Alan?

Yes, we did when we announced our fourth fund. We announced this in February, and we talked about how one of the reasons we were able to raise this fund at a time when many were struggling—not everyone, but many—was that we had a good return.

The reason we had a good return is because we always believed that the main use of blockchains was finance, so we focused on that. Because we really focused on that, we were able to pick better winners and have more of them. For us, it feels like a continuation of the same thing, at a time when I think a lot of other people are trying to figure out what the future is going to look like.

I say this because, looking back over the last 18 months, I find the future a little more murky because of what's happening with AI and how it's changing absolutely everything we do. But the foundation of what we invest in, the foundation of what our funds look like now, is DeFi, CeFi, and the market infrastructure—the crypto infrastructure that enables these things—stablecoins, payments, and applications that create new financial products based on blockchains.

I think about the investment in Polymarket that we made a little over 2.5 years ago. I remember recently reviewing a memo that I wrote with Omar from our team, and we talked a lot about how social Polymarket was, and how it was the combination of the social aspect with the financial aspect that was going to create such a breakthrough consumer app.

I still believe that this is true, that there will be more and more social things. You see, for example, FOMO, which raised a bunch of money, has a chat feature, and Polymarket has a chat feature, but it's still finance, and everything is becoming financialized. It's not strictly a crypto moment, but financialization is the foundation of how Generation Alpha and Generation Z interact with the world.

So I expect that, when I think about consumer applications, that should still be the future. Having said that, we continue to think about things this way. We continue to be very focused on things with a financial aspect. But I really think that finance and AI are now coming together in all these different types of things that we used to call consumer.

Yeah, I think if you take crypto to its logical limit, it leans more toward an HFT-type infrastructure, to me—not 1-to-1 in the case where it's a focus on nanoseconds and picoseconds, but generally just a play to increase trading volumes.

And to your point, I think even when Robinhood was just starting out, it was like, “Okay, you're going to do zero-commission brokerage and you're going to serve retail clients.” Retail—there's no flow there. Now hedge funds are watching WallStreetBets because you can get knocked out of a short position.

5. Blockchain as the New Settlement & Issuance Layer

So I think I'm definitely playing for on-chain volume to increase, and I'm glad it's starting to pick up momentum. It seems like there were a lot of issues with scaling, especially in the early days, like the DeFi Summer and NFTs, when you were paying $500 or $1,000 for transfers. But now, as blockchains become more performant, we can do more trading and develop infrastructure.

We seem to be moving from NFTs and memes to real stocks, derivatives, and currencies, and it looks like more and more things are coming to blockchain.

Yes, I agree, that's quite right. You mention the Robinhood example. Options are the largest revenue stream for Robinhood, and 70% of the volume of these options are options with an expiration date on the same day, right? I don't think many institutional players trade these options on Robinhood.

This says something about how retail investors feel about trading today. Robinhood's primary user base is no longer Generation Alpha or Generation Z; its primary user base is millennials. It's like me: I've grown up on Robinhood over the last 15 years, and that's where I started my trading business.

There used to be a mindset that, as I got older, I would switch to Schwab, and I did, but many of my generation didn't. They just stayed there, and that's why Robinhood now has cards, asset management, and so on.

It is clear to me that blockchain trading activity will grow; there is no doubt about it. We're in a strange situation now where they're talking about stablecoins and having atomic settlement, and that creates a bunch of different problems. All real-world payments currently taking place on the stablecoin network are still processed in batches, and many tokenized securities transactions are also settled in batches—once a day, during banking hours.

So I have a question about this gap. Current software systems, accounting systems, the way brokers work, and regulatory infrastructure still can't interface with this 24/7 market that runs on blockchain. While many are starting to build infrastructure there, and I really think that's the logical path forward, especially when we're facing a lot of regulatory resistance—moving from legacy systems to working online—I think there's still a long way to go.

This seems to be mostly a regulatory issue, but if we go back to the infrastructure aspect, I would say that no one has gained a definitive advantage yet. There's still a lot of room for experimentation with payments or trading, and it seems to me that even if we move toward creating a universal or global exchange, the regulatory problem is not just limited to the US; it's potentially global. So I'm very interested in watching how things develop.

Uh-huh. I mean, the biggest winner is definitely the United States, right? Because we're just exporting US dollars and stocks to everyone else.

That's right. From an infrastructure perspective, you're absolutely right. Payments and trading are a bit different in this case, but even all the big crypto payment companies are, right? Almost all of them—not all of them, of course, but almost all of them—were exchanges that changed the vector of activity.

The infrastructure is actually still quite weak compared to what one would think is necessary for real movement of funds. One of our theses is that many of the first group of companies that are currently causing excitement are unlikely to succeed over time because they have significant technical debt that is not aligned with the future we are all talking about.

The second and third waves of entrepreneurs coming into this field are from the largest payment companies in the world, who really understand the movement of capital better than that first group and are structurally building much more logical systems for the future.

The same thing happens in the trading space, although many of the early crypto adopters were people from Citadel, HRT, and Tower, so they approached it a bit more carefully. I believe in the success of quants. If we're moving toward trading and blockchains are mostly finance, then companies like Citadel, Jump, or Jane Street will be able to adapt quite well.

But I'm curious. One idea that I've been thinking about a lot from an investor or venture capitalist perspective is that, if the blockchain space is primarily about finance, then I don't think there will be thousands of completely new primitives, as most people thought when comparing Web2 and Web3. We now generally understand finance; we understand payments and trading.

Payments, trading, and finance in general may grow 100 or 1,000 times on the web, but that doesn't mean there will be 100 or 1,000 times more new primitives. The space could grow 1,000 times, but from a venture capital perspective, I think the number of developers could be smaller, or the ideas might already be known.

The question is who will capture the greater market share. Perhaps it will be a crypto-native project that already has market share and will simply continue to grow. But I'm curious what you think about it.

Yes. I think that if stablecoins and tokenized assets become something like water, they will be everywhere, right?

Mhm.

And I think that's already happening. If they become part of everyday life, they will be everywhere. Everyone should drink water. If everyone has to drink water, then we have a huge funnel for all these different things to build.

Does this mean that 1,000 new primitives will appear? Not necessarily. But we have the largest market in the world, namely finance. In my opinion, we are in for a revolution in the modernization and digitalization of finance.

Therefore, the number of things or good companies that can be created for this is simply huge, and the possibilities are many. We also now live in a world where building software is easier than ever, and thanks to blockchains, fewer intermediaries are needed.

So we are definitely—without a doubt, and it's happening in the AI space, but I think crypto and blockchain are part of it—the biggest companies of the future are going to be much bigger than the ones of today. It's not just about inflation; it is also the ability to cover a larger part of the value chain because we have fewer legacy systems.

I see this as a huge opportunity for potentially great founders and huge markets, but probably in less specific verticals or primitives. It also means that, if you look at payments, for example, there's Worldpay, Adyen, Checkout.com, Stripe, and a bunch of companies that are worth $50 billion to $200 billion—and that's old-school payments.

In my opinion, in the world that I'm interested in, there will continue to be a lot of really big opportunities and companies. There's also a globalization of all markets, which again means the creation of bigger companies.

One of the strange things that was true in the crypto world was that we would take a product, like, “Here's a structured product, one deal,” then put it in a DeFi protocol, throw $6 billion in TVL in there, and say, “This is a colossal business,” right? If it weren't for this strange problem of limited supply and demand for tokens, it would never have become a business in its own right.

And we're probably going to come back to this in the crypto industry, which means we have to think carefully about how big a market this is. One of the things we talk about a lot here with the investment team is what games we play. Are we playing big enough games? Many of the games the crypto world has played over the past 10 years have been relatively small, at least in terms of individual investments, even though the ideology behind it all has been massive.

I'm optimistic about the future because I think the chances of creating a company worth $100 billion, $200 billion, or even $1 trillion are much higher now than before, but the amount of knowledge I'll have to master is probably less than before. I just need to dive deeper into these specific markets.

Yes, I think that's quite logical. So this is kind of the next stage in the evolution of fintech in general. Hmm, interesting. How do you generally look at it when it comes to finance? I believe there are many similar elements, but of course it won't be a complete 1:1 correspondence.

6. Prediction Markets, Information & the Next Interface

I think payments is a good example—or even Stripe. Well, not Stripe specifically, but credit cards in general are a good business because they usually charge 2.9% plus a small transaction fee. By the way, I worked on this at Tesla because we were spending a lot of money on credit card fees on the Supercharger network, and we were thinking, “How can we reduce these costs?” But if you transfer this to blockchain rails, without cards, then transferring stablecoins would cost only a fraction of a cent.

So how do you generally assess the transformation of the business model with the transition to blockchain platforms?

This is a valid question, and one I get asked a lot, especially with Ring, since they offer these cards. My general opinion is this: people often forget what exactly they are paying for in any payment system—fraud protection.

Yes.

Fraud, compliance, and deep integration—many of these companies provide equipment for free, so there's a lot of cost built into these things. These programs, when you talk about basic programs, require a huge amount of work across the value chain. So there's a reason they cost something.

We don't live in a world today where P2P transfers of stablecoins could in any way meet the needs of, frankly, anyone on a real scale. You have to be very, very savvy to be able to do a P2P transfer of stablecoins. Even you and I sometimes make mistakes, like sending a test transaction, which obviously doesn't work in the long run.

The truth is that we're in New York right now, and there are probably about 200 points of sale around us within a 3-block radius. They all already exist, with existing independent software vendors, or ISVs, existing software, and payment processors. So I think it will be very difficult to displace what already exists and to embed into these existing processes when you don't already have an existing network.

But I really believe we will see changes in payment channels and in many banking networks. That's my own thesis, and we've expressed it in the way we invest. I know a lot of people disagree with this, but I don't think you're going to replace Visa—or at least I don't think you're going to replace Visa in the payment flows for a lot of these banks.

I don't think you will replace the anti-fraud and compliance work that is currently underway. But I really think what's going to happen is we're going to get rid of the T+2 system. For Visa, if you use issuer-side stablecoin settlements, it takes maybe T+8 or T+12 hours because they can only really do it once a day. Eventually, it could become real-time, and it probably will, but that's not the case today.

Later on, it will also be possible on the acquiring side, meaning the other side of the transaction that concerns the merchant. Eventually, these merchants will also have wallets at their points of sale and in their apps, and they won't even know that it's a stablecoin. Suddenly, you have an existing authorization network and existing anti-fraud and compliance standards, but you just changed the way all the calculations work.

You got rid of the excess and became much more efficient in using capital. You got rid of a lot of the need for excess capital in banks. You've gotten rid of some of the need for certain types of fraud and compliance. You still have some financial fraud, but not all of it. You made it much cheaper, made it more real-time, and allowed for a much more global ecosystem.

Some of this depends on how you look at exchange rates over time. Maybe I'm a dollar maximalist, but I think that's what the world is heading toward. I think we're just going to reinvent everything with more modern rails, more modern banking and core infrastructures, reinventing the way payments are made.

So you think that companies that implement this simply make higher profits, right, because they're just more efficient on the backend?

Yes. If I think about the credit-card transaction fee now for Tesla, whatever you pay—that 30 Visa points—and you pay interchange, that's the bank, and maybe the ISV and the processor there. All together, it's 3–3.5%, right?

There's really no incentive for anyone in that chain to get rid of that commission if you can otherwise shift or distribute the economics. A lot of what's happening is not getting rid of these 3–3.5% fees. It's just a different distribution.

For Tesla, or any company, they say, “We would like to get rid of these fees.” But what if I said, “The commissions still exist, but Tesla can have a revenue share and get a piece of it”? Then you can pass the rest on to the consumer. Or maybe the commission is still there, but you could potentially just give a discount to the consumer, which wouldn't be relevant for Tesla.

If you think about some neobanks or some money-transfer providers and so on, that's something they would obviously want. To me, it looks more like, yes, we will see fees come down over time, but it's more of a redistribution of fees than anything else. It's probably going to have the most negative impact on banks.

Would you say that inter-network interaction is following a similar path? Because if broader trading moves online, I think that currency and other markets have historically—and correct me if I'm wrong—had pretty high spreads, at least in airports. You are simply being ripped off at the airport, but it seems that there are potential opportunities for efficiency improvements in general and in the retail sector.

Yes, I think that's absolutely correct. A lot of the big liquidity providers that people think of in the context of payments and all kinds of trade and settlement that we're seeing right now—B2C2, Wintermute, Nonos, FalconX, and so on—are looking for local-currency providers.

There's a lot of talk about how we need tokenized currencies to work with stablecoins, right? If we have this end-to-end ability to do on-chain transactions, it becomes much cheaper, because right now there's a lot of friction and costs at these endpoints.

You need a 2-way flow for this. You need to be able to net the flows, and in U.S. dollars there is a large 2-way flow to all these countries and back. Frankly, there is no 2-way flow of Canadian dollars. There is only a 1-way flow of the Canadian dollar, and because of that you need fiat currency on the other side.

What is needed, I think, is that forex services will be significantly reduced. Those liquidity providers that you mentioned at the beginning are essentially trading platforms that are looking for better ways, like interbank, multicurrency platforms. That's why I think the foreign-exchange market is very interesting right now, because it has always existed in banks, and now more and more things are happening outside of them.

I think it will happen in a more digital way, but it will still probably remain fiat for a long time. This will not be tokenized until there is a significantly larger volume of tokenized trading across various markets. While I see a logical conclusion, it's unlikely to happen in the next 10 years.

So it's more of a redistribution than a complete disappearance of something.

Yes, that was generally my point, and I think you can see that there is pressure on the commission right now. We are the new intermediaries, in a sense, and we democratize intermediation because we can share the profits with the end consumer.

This is what we see from the neobanks. All the neobanks say, “We'll give you 3%, 4%, or 5% cashback.” Well, how exactly do they subsidize it? They do this through tokenized returns and interchange. Instead of the issuing bank keeping it for itself, it shares it with the client.

Interesting.

Even with stablecoins, Tether has a real printing press thanks to Treasury bonds, but Agora and others are trying to return this in one form or another to end users or partners.

So, again, maybe it's just massive redistribution. If you think about Tether's business model over the years, it's an incredible scheme: I borrow a dollar from you—you essentially lend it to me. I give you a debt in return. Then I invest these funds in the U.S. Treasury, and I don't charge you anything for it. So in terms of real-dollar inflation, you remain in the red, and I take 4.5% for myself.

True.

And so you see that, yes, there is pressure on the commission. But in reality, my bank deposit now works as if Circle, Agora, Ethena, or another stablecoin issuer were sharing the profits with me, or as if a neobank were sharing them with me. I invest money in, say, Plasma 1, and it pays me the income from the stablecoin. In fact, all that happens is simply a redistribution of funds from the bank to the end consumer.

I think the main theme of what’s happening at Rain, with interchange and stablecoins, is that we’re trying to redistribute fees from the bank to the consumer. Ultimately, this is what people are talking about: “What does this mean for lending?” Banks are the biggest lenders in the world, and this is still an unresolved problem.

7. Crypto VC Consolidation & Dragonfly’s Thesis

But I also think that if you look at how the world has evolved, even outside of crypto, who has become that margin lender? It became insurance companies; it became private credit funds, right? And so it seems like we live in a world where the bank is breaking up into all these different pieces. This is something like the unboxing of the bank, which I talked about in fintech 10 years ago, and it is generally good for the end consumer, although there are still issues with loans.

So, we short banks, long crypto.

Yes. Well, and that’s why all banks should try to implement tokenized assets and stablecoins.

Interesting. I think, overall, one of the main unresolved issues in the crypto market has become AI. Maybe not just AI, but also some growth in speculation beyond crypto or just tokenized assets. For a while, crypto was the only really new technology—at least one where regular people could speculate—and that’s probably still the case today, but technologies outside of crypto have become very real.

Perhaps there’s a brain drain, so to speak, not only of dollars but also of talent. So it feels like a barbell approach: long on finance, long on trading, and, on the other hand, long on AI, potentially long on robots. I’m optimistic about robots, but maybe it’s a little too early.

But how do you think about it? Even at the conference we were at earlier this week, everyone was talking about this intersection. Do you think there’s a lot in common between AI and crypto, or do you think they’re parallel paths that will develop separately?

Listen, I think that, based on what I was talking about earlier, tokenized assets and stablecoins becoming part of the system means that they will intersect with all the new technologies. It’s also very clear that AI has become part of this system, right? It will permeate everything we do, and so there will be a lot of intersections because of this.

But does this mean that a lot of what people are talking about at this intersection will come to fruition and bring a ton of value to protocols and token projects? Honestly, probably not. There are specific companies and protocols that are becoming better because they can use crypto. There’s a lot of talk right now about how agents pay each other with stablecoins. I am absolutely certain that one of the payment methods for agents will be stablecoins and blockchain networks.

I really think programmable smart contracts are a good foundation for agents to work on. This is true for tokenized assets and stablecoins, but does this mean I should invest in altcoins? Apparently, these are not related things at all. Do you understand what I mean?

And I say this because when I talk to my companies and our protocol teams, they need to realize that we live in a world where most of the code next year will probably be written by artificial intelligence, not humans. So when you think about building infrastructure for your business, you need to understand that your API will most likely be interacted with by AI, not a human.

This changes your approach to infrastructure planning in the future. It changes your vision of the programmability of what you do. It even changes how you perceive issues of fraud and compliance. But does this mean that everything should move to public blockchains? Probably not.

I was initially inspired by the potential intersection in the DePIN space, but pretty much all the crypto companies that started with AI and found their market simply switched completely to AI. They can use crypto as a backend, but more in the sense that we need banks and infrastructure, so they can use it as financial rails or just for the movement of stablecoins.

But it’s a little frustrating that none of them really need crypto to create a better product. They use it because it is available, not because it is at the intersection of these technologies.

They make each one a little better.

Yes. Many early crypto-AI projects had essentially the same thesis as DePIN.

Yes, that’s right. It consisted in incentivizing people or systems to perform certain actions using tokens with high emissions: They do it, the project becomes large enough, and then the emissions stop.

The only project that managed to reach second space velocity was Helium, but that was due to its hyperinflationary tokenomics and the fact that it was 2021. Everyone else who tried to launch a high token inflation schedule simply brought everything to zero.

Yes. And even Helium—look what happened, right? They sold off the bulk of the business, which was doing really well. And I think if you asked them what they thought about how the tokenomics was designed, they would say it was a mistake.

The same thing is happening now in decentralized learning or data labeling; the same thing is happening with graphs and so on. No one knows what the final state of these protocols and businesses will be, but it seems very similar to me, and I expect that for many of them, things will follow the same scenario.

So I became very skeptical about financial engineering. I’ve seen so many clever financial schemes that just blew up, so now I’m saying, “Okay, show me a better product. Show me how you’re going to make money.” And if you say “financial engineering,” I just don’t want to deal with it.

Well, financial engineering is a valuable thing, and it’s valuable because you can find inefficiencies and build something around those inefficiencies, right? Take MicroStrategy: if you look at it as financial engineering, you look at it and think, “Oh my God, what they did with Strategy is a big problem for Bitcoin.” But it wouldn’t have become a $100 billion company or whatever. It’s a $100 billion company now, and if financial engineering didn’t exist, it wouldn’t be like that.

And you know what? We wouldn’t have a private equity industry if financial engineering didn’t exist. But the point of all this is that financial engineering means that you have noticed an inefficiency, and that arbitrage will disappear over time.

So you have to benefit from this financial engineering until you are unique or until other people have filled this arbitrage niche. Then it disappears, and you need a product. So you can use financial engineering to achieve product-market fit, but actually, that’s what DPIN was trying to do. I’m probably preaching to the choir, but you have to get to product-market fit and the revenue model.

Perhaps many DPIN protocols misunderstood that product-market fit is not just about users; they must also be willing to pay for it.

Yes. And you need both. You need monetization and users. Yes. It was difficult with DPIN. I’m a little sad because I feel like the idea was interesting, but it seems hard to find even one DPIN project, other than maybe Grass, that potentially makes decent money.

Even from a token perspective in the broadest sense, the market—and perhaps this applies not just to DPIN in general, but to tokens as well—has simply not yet rewarded revenue, at least historically.

Going back to your market opinion, they all traded the same, with the beta to Bitcoin. Hopefully, over time, as we become more efficient, these tokens will be more dependent on revenue and product-market fit. It seems like that’s already happening a little bit now.

If you look at the best-performing tokens, they are Hyperliquid, many trading applications, many stablecoin-related projects, and some revenue-generating AI projects. So it feels like we live in a world of dispersion, doesn’t it? And dispersion is good because it means the market has become more efficient.

Well, or at least it’s good from my perspective. Perhaps those who benefit from the lack of variance don’t think so. But we’re definitely getting to the point where, while we still talk about crypto as an industry—and it certainly is a separate industry—we’re gradually moving to a state where the entire crypto industry isn’t necessarily doing well or badly. Are specific companies that build using crypto technologies doing well or poorly?

By the way, it’s just technology. We used to have Internet VCs, remember? Today, no one says “Internet VC” anymore. Everything is the Internet, right? And I think that’s probably what’s going to happen with crypto venture capital. I think that’s probably what’s going to happen with AI venture capital eventually, because everything is going to be part of that.

I agree. And we actually often have this conversation with LPs in general. This is how crypto becomes more like venture capital. One could argue that this is like a fintech venture, but now companies actually have to achieve product-market fit. You can’t just launch a token. And that is optimistic. This means that there is much greater variance.

I think you said a while ago—or recently—that they are like a meteor shower for crypto venture capital investments.

8. Real-World Assets, Stablecoins & On-Chain Markets

Oh yes. And largely because I would say that many crypto ventures have historically not played the “best product” game.

They were just playing a token game. When tokens are listed on Binance or any exchange and then immediately drop in price, that game is over. Many people played this game, but the rules have changed.

Well, I don't know if you do, but we do. There are many companies that do venture-capital benchmarking. You provide them with your data, and they send you anonymized data in return.

Yes. Right.

We see this both for ourselves compared with other crypto venture funds and compared with broader venture funds. What you notice—and you can also see this by looking at the Carter data—is that the DPI and TVPI of crypto venture funds are still significantly higher than the median of traditional venture funds.

That was the story of why it worked. LPs said, “Okay, I'm taking on all this volatility, and I don't know how to talk about most of the things in your portfolio. It doesn't make sense to me, but I get faster DPI, and it seems like I'm making more money doing it.”

And that's now disappearing because of how tokens are behaving. A lot of those venture capitalists were saying, “Okay, I think this is...” I remember people saying that all the time: “I don't believe it, but I think this token will show good results. This token will grow a little. I can get out.”

The token is the product.

Yes, a token is a product. This allowed people to make money, but I don't think that will continue to be the case. If that's no longer the case, then a significant portion of these venture capitalists will simply die out, right?

There is already one large and very well-known venture fund about which there are quite credible rumors. I've heard from investors that they're telling their depositors that they're not going to raise a new fund. This is a brand that everyone knows. They may be the first, but they certainly won't be the last.

If the incentives are such that the founders of that fund have already made enough money, and if they have to play a completely different game and reinvent themselves, making it much more difficult, then maybe they're not interested in it anymore. They can do something else, like robotics or whatever—or lie on the beach.

Yes. In the long run, it's great.

I spoke to a fairly well-known investor who has invested in many, many funds, and his conclusion was that while crypto is maturing a bit, people who are focused on better products—with, I wouldn't even call it early liquidity, but the difference between stocks and, say, an acquisition or an IPO versus a potential token issuance—still have a higher TVPI and possibly a higher DPI than the long-term paths through an IPO or acquisition.

The token typically acts as an instrument, and the tokens that have performed better today are the ones where there is revenue and product-market fit. Overall, the rules of the game have changed dramatically. For us, at least at Frictionless, we've always tried to invest in better products. The technology was cool and I liked digging into it, but ultimately you have to find product-market fit. You have to be earning revenue.

Of course, no one would bet on 1,000, but overall, that was our aspiration. For a while, it seemed like we were crazy. I come from the grocery industry, and that's why I say: create good products. Things are stable now, and as you aptly noted, there is some dispersion among the allocators of capital.

Yes. Certainly, we see that the world of crypto venture funds is shrinking, and the data supports it. Crypto investors now say, “I'm a venture capitalist, not a crypto investor,” or “I'm an advanced-technology investor,” or something similar.

This brings us back to my point about “internet venture capitalists”: no one calls themselves that anymore. I think it will be the same here.

From the perspective of other allocators, if you invested in, say, an early round of Lightspark, as we did—this is a disclaimer—then you still got potentially better, easier, or faster liquidity. We didn't sell anything, but some VCs can get liquidity faster than if it were Robinhood, right? There is such a story.

I expect that, in the absence of clear legislation, the situation will probably change significantly. I expect people will continue to work hard on how best to accumulate value over time. If we get clarity and people start to feel positive about the possibility of tokenization and replacing capital with tokens or something like that, then other venture capitalists—not just crypto investors—will eventually come into this niche.

In 2021, they all came in and tried to figure out how to work with custodial services and things like that. Now it's becoming much easier to secure and control assets. If this becomes the main way to return capital, then everyone will understand how it works.

So, theoretically, if we succeed—again, subject to legal certainty—all companies will become, so to speak, public or tokenized on the blockchain. At least, that's what we hope.

That would be very interesting, maybe in essence, and we can wrap up after that. A lot of funds have already entered the Frontier or robotics space. In terms of robotics, I feel like it's similar to the ChatGPT moment in early 2022, when people were saying, “Oh, this is cool,” but it was really only with Opus, version 4.5—and, at least for me, in the last year—that people realized, “Oh, AI is really a big thing,” and started investing in it seriously.

It feels the same with robots, although people generally still underestimate it. If we move forward 10 years, there will be hundreds of millions of them. So I'm curious: as far as I know—and correct me if I'm wrong—I don't think Dragonfly has gone beyond the cryptosphere.

We didn't go out. To the extent that we're a fintech fund, we have projects related to AI and so on, but I think we're very deliberate about where our advantage lies.

One of the things I always say is that when a deal comes in, you really have to ask yourself 2 questions: Did I see it first? And if I didn't, do I have some special ability to appreciate it? If those 2 things aren't true, then you shouldn't make the deal, right?

I know personally, and I would say this to everyone at Dragonfly, that we don't have a particular ability to evaluate robotics. Besides, we definitely didn't see any deals in robotics first. I can confidently say that this is also true of some other venture capitalists who say, “Oh, I'm going to do robotics.”

I still think that crypto was niche and innovative at the time, and perhaps we're now moving up the S-curve toward mainstream adoption, now that we've focused on finance, commerce, and payments. I don't know. I think there's always an opportunity to learn, but I was wondering why you guys are still mostly focused on the crypto market while some others are expanding.

I think there are a few points here. First, the crypto community is usually pretty good at first-order thinking and at thinking from first principles, because we were forced to do it and because we approached it from a rather strange angle.

A lot of what's happened in the last 8 years since Ethereum came out—or maybe 10 years now—is that we've been making the same mistakes that finance and Wall Street have been making, just sometimes in a different, weirder, more absurd way. It was quite interesting.

I think it really depends on the context and the perspective you're looking at it from. I personally have investments in defense technology, artificial intelligence, and robotics because there's a lot to learn, and I think these are big trends. I have funds that I think are worth spending there to diversify.

But as a fund, I have a fiduciary responsibility to large, well-known LPs who ask me to perform a task for them that they cannot do better elsewhere. I'm sure I can't do the robotics task any better than some of the other managers in the market, and I want to act fairly toward these people and make good profits.

I'm also motivated to make good profits. We are our own largest LP because we invest in our own funds, and I hope most GPs do the same. That makes a difference in how I think. I can say, “Oh, that's interesting to think about,” but, to be honest, for a lot of my personal investments—especially when it comes to private things—my vetting is less rigorous than what I do when choosing which bagel to order in the morning.

9. Closing Thoughts: Where Value Accrues Next

On the LP side, the scope of my vetting is very deep. We're known as a fund that dives very deeply into working with companies because I have a different responsibility for that capital. I'm meeting a different need and responding to a different request from those LPs.

Right. Probably the last question. Regarding the crypto community, what are your current thoughts on ETH? Is it done? Is the “value team” dead?

For the record, we were criticized when we raised our fund. We said we weren't optimistic about Ethereum. We're generally not very optimistic about L2s. Mostly, we just focus on other scales.

I said ETH was like MySpace in 2022, and everyone was like, “You're a [ __ ] idiot.” I don't know. It was quite interesting just seeing how things continue to change over time.

But what do you think about some of the original blockchains?

Listen, it's hard to say. I still personally own ETH, Bitcoin, and Solana, so I'm not saying these things won't increase in value. Given everything we just talked about, I expect a lot more activity on the network in the future.

Of course, from a value accumulation perspective and our approach, if we believe that these assets accumulate value based on income multiples and so on, it's difficult to estimate how much they'll be worth in the future. But I also think things are very rarely traded based on fundamentals. Honestly, I think it's all about momentum. If we go from, say, X transactions to X plus 1 million transactions, then everything will grow, and I'm optimistic about this space.

I believe in financialization on blockchain rails, so let's not get rid of a store of value anytime soon. No. You can't refuse it. Class.

Well, thank you, Rob. Appreciate it.

And thank you, Logan.