The Man Behind hundreds of Web3 Investments: "Tokenize or Die" | Yat Siu

There's nothing that won't be tokenized. We have a saying here in the company: tokenize or die. In the next 10 years, there will be over 100, maybe even 200 billion AI agents, and basically everything will be transacted by AI agents. Then dollar colonialism is very successful, and in the end you're basically conquered by the US dollar economy. And I think that risk, especially for Europe... we were one of Apple's biggest app developers. Apple didn't like something we did.

Well, I think my parents didn't even know what I had done. My parents were already divorced when I was much younger, so I basically only lived with my mother. And she was also traveling a lot. My mother is an artist. She was first an opera singer and then did directing work, and she was always on tour. So actually when I was 13 I was probably home alone most of the time. And nowadays I find that kind of hard to imagine. Probably child services would show up and say, 'Okay, what's going on here?' But back then nobody cared at all. And she only knew that I had a computer, and back then I had an acoustic coupler first and then a modem, and I used that to upload things and I just had things to do. So I think she only understood much later that I was actually in the computer industry, so to speak, because that wasn't really a subject back then. It was like data processing. Nobody knew what that was. At least my mother definitely didn't.

Yeah. So it started in the early 80s. My first computer was a Texas Instruments TI-99, that was really like a huge calculator, right? Like with 16 kilobytes. And then I started programming in BASIC. It was just fun and relatively simple, and there were no courses either. After that with Commodore and then with Atari. And I think what happened there, that was in the 80s, that was before Windows, and computers were really pretty new things. But people also thought back then that when you buy a computer you'd have it for the next five or 10 years. It was like a car back then. But now of course it's completely different. But what was also interesting is because there were so few people who could actually handle computers in general, people didn't really care how old you were or where you were from. Because in that sense, you actually needed talent. And you see that nowadays too. Not just with the internet or in crypto or in gaming or in AI. If you have talent, it doesn't matter how old you are or where you're from, and you can get a job anywhere. And that's basically what happened to me too.

Right. That's a difference of almost 30 years. But what happened was I actually studied music first, and then because of music I programmed MIDI software that got me a job at Atari. They sent me to America. And in America, Atari went under during the early 90s. But during that time, of course, the internet as we see it now with email and websites, that was the era of Mosaic and Netscape and the growth of Java and Sun Microsystems, that was all there in the early 90s. And I basically just saw what was happening there with the potential that was coming. And it was always about, back then it wasn't just about information but generally a decentralized information system. Meaning, in the 80s and let's say also most of the 90s, information was really something only for the people who had the privilege. Did you have an encyclopedia at home or did you have access to many books? Did you have access maybe to a TV? Those were things that weren't typical. So only certain groups could have that and then had information.

And with the internet, it became so widespread and you could really bring the masses into it so they could have more information, the same information that everyone else had. And that was how this information arbitrage went away. And because of that, we thought it's better for democracy and better for the world and more opportunities. And that is indeed what happened. But then in the meantime, because of the data and the need to bring all these data centers and computing power together, the data consolidated with the big companies: Facebook, Google, Microsoft and so on. And what was actually valuable about this so-called free open information that was everywhere was actually the data that we as humans then provided. So it wasn't just information we could read, which was the first version of the internet, but also the internet where we contributed. In forums first and then in blogs and then in websites and then social media. And we basically became the biggest data producers for these platforms, which learned more information about us, more information about the world in general, and became very powerful and basically used this information because it was in their own databases as their own property, even though it wasn't really their property because it was our information that we contributed, and created this closed internet. That's Web 2, and we're generally still in this phase.

And how we got to blockchain, that was really, we're talking about 2017 here. We had various experiences. One of them was we were one of Apple's biggest app developers. Apple didn't like something we did, but they didn't call us and say, 'Hey, can you change something here?' They just took away all our apps and our whole business on Apple simply because someone back then in Cupertino said they don't like us. And that was the internet where we were all of a sudden subservient to someone. It was no longer the free internet as we thought of it back then with open source and Linux and the first generation. And what we then saw with CryptoKitties, which was our first big experience with blockchain, with NFTs, was the possibility that this data and information was no longer controlled by the big companies but was actually user-generated, user-owned, and the supply and demand belong to the user and was totally open on a data system that basically belongs to everyone. The blockchain is the public data system. And that was one of the bigger reasons why we went really deep into the blockchain story, because we saw the potential there to actually bring the internet back to the free decentralized internet which was actually taken away from us by the first phase of the internet.

Yeah, so generally I was active very early in the internet space. I founded one of the first ISPs, an internet service provider, for network services and so on. And after that, in '98 I founded a company called Outplace. I had sold the previous company, still wanted to be in the internet space but didn't know exactly how I could compete with big companies like Yahoo back then. Yahoo was a big company, or GeoCities, or AOL for example. And I thought, what can I actually do? And we became a sort of picks and shovels company for people who wanted to get started on the internet. And we had all these products like email and websites and calendar services. So like a sort of Google Calendar, Google email before Google, which we then sold to companies as a service provider. And that became relatively successful. We had about 90 million users at our peak, and that was back when the internet was still relatively small of course. So we had a pretty big share. I think almost 20 or 22% of all emails in the world went through our system. And then in 2009 IBM bought Outplace and integrated it into Lotus Notes as the cloud version of Lotus Notes, for the people who are old enough to remember what Lotus Notes even was. That was the office software back then, the big competition from IBM against Microsoft Office back in the day, and was actually a pretty huge business for IBM back then. After that, of course, we also had a gaming, video game department in the company. That was the founding material for Animoca, where we actually built mobile apps. That was around 2009, 2010. And in 2011, we had our first big video game hit. And that was actually the founding of Animoca as it is now, as an original video game company before we went into blockchain.

Yeah. 2011 normal video game, we were a mobile games operator. So that wasn't a normal video game back then. Normal video games back then were PC or console games. Mobile games were very new back then and grew rapidly because of that, and that was also their time. So in 2012, Apple took away all our apps. We then became relatively big on Android. It was called Android Market back then and now it's Play Store. But it was a difficult time because we were market leaders at first and then we lost everything because of Apple, and then years later we brought our game company back but obviously not in the same format as before because we had a disadvantage. And one of our studios in Vancouver then co-created a game with another company in Vancouver, and it was called CryptoKitties. And that was an experiment back then. It was an experiment to see how you could build blockchain technology on this non-fungible system for basically digital cats. And that was just a bit funny, a bit of fun. And of course as an experiment, which was originally a hackathon, what happened then was still very early days while Ethereum was still relatively at the beginning. The trading of these digital cats was about $40 million within a week or so. The Ethereum blockchain couldn't scale. It basically just collapsed.

And that made for interesting headlines like 'blockchain can't even handle a few million dollars.' And on top of that with digital cats, which was a bit funny, of course, but at the same time a bit scandalous. And it was also very controversial because a lot of people in the Ethereum blockchain didn't want that. But for us, that was proof that it wasn't just an interesting technology where you could actually have digital property rights, but also that there was a demand to buy digital cats, which was also interesting. And from there, we went into the world of blockchain. But our start was through NFTs. We didn't go in through Bitcoin. Bitcoin has a much earlier start, like 2009, 2010, and we came much later to that because Bitcoin has a very strong financial concept which wasn't that interesting to us back then. And I think that was also one of the concepts we understood later: that even if you're an entrepreneur and in business, it doesn't mean you're necessarily interested in finance, even though it's incredibly important. And it was these previous experiences that then provided the information for what we actually wanted to do as a company and what we thought was important for the world. But we started through digital cats with CryptoKitties.

Yeah, I mean an NFT is also a token, just like Bitcoin or an Ethereum token, like a SAND token or whatever tokens there are. But the difference is it's non-fungible, hence NFT, and non-fungible really means unique, there's only one of its kind. And I think the best way to describe it is because it's verified by blockchain technology, that means you know exactly you can't fake it. It's basically the original, or like a certificate. For example, if you buy a Rolex watch you get a certificate for it, and this certificate says, 'Okay, we say here that this is the unique one, and this is number 300 of this watch, and this is the proof of that.' And that's what blockchain technology can do as an NFT. It says, 'Okay, this watch or this object or this digital object is unique and you can prove it through the blockchain,' just like you can prove that if I send you a bitcoin or receive a bitcoin that it's the real bitcoin and that it's not forged and that it's not reproducible. So it's the same technology just to prove unique objects rather than fungible tokens like Bitcoin or Ethereum where every token is the same, not just in value but in meaning.

Yeah, and I think it was different things. We weren't just involved in Sandbox, which was one, we were practically part of the founding team, but also with Axie Infinity and with F1, and we're also involved in Pudgies and also in Bored Apes. So we were involved in many of these projects. And one of the big reasons why we did that back then was the thesis that video game users actually in the end want to own the objects they buy, because skin trading, which are digital goods that aren't on the blockchain but are already traded, like Counter-Strike skins, are now worth billions in terms of market value. They trade in the billions of dollars, but that's all controlled by a central company that always changes the rules. So not a real market. And with blockchain, because you actually own the object digitally at least, you can actually make it a real market. Because if I can't buy it there, then maybe I can buy it somewhere else. Just like if I had a physical Pokémon card, for example, and I wanted to sell it, I wouldn't necessarily have to sell it on just one platform. I could sell it on eBay or Amazon or sell it directly to you, send it to someone and sell it directly, because it belongs to me and I can do whatever I want. Whereas with video games, you didn't actually have any ownership rights. And we thought back then, and it's still part of the thesis, that in the end video game users would like to keep their objects. Not necessarily that they want to sell them, but that there's a certain personal value that isn't just represented by money, but also by status and reputation, what I've done in the last 10 years, things you collect at home, for example, because it has a personal meaning. And we had just invested heavily in this thesis, which now of course was also a big reason how Animoca grew. And in the meantime of course that has evolved, because for example NFTs are now mostly used as technology for Pokémon cards, although it started with digital cats and of course with Bored Apes and Pudgy Penguins and so on. The main trade with NFTs is now with actual Pokémon cards, because every single Pokémon card, even though it might look the same, is actually unique because of its age or how it looks. And now the Pokémon trade has grown significantly because of NFT as a technology. So that's kind of where the product and the technology have matured. And you see that generally now in the whole crypto blockchain scene, where technology that was experimented with on memecoins or other things is now being brought into the financial world for very serious things, let's put it that way.

The space has evolved. The thesis hasn't really changed. So just for context, when we started in 2017 and 2018, I think the entire crypto scene with NFTs as well as with tokens was maybe $500 billion. That was the total market value. In the meantime, we are now also with Bitcoin, from peaks, now of course it has dropped. So we are now at 2.2 or 2.3 trillion. And so from that perspective, now we see that over the last 10 years the market has risen sharply. It's just relative if you say, 'Okay, how was the price a year ago versus this year?' It's lower, that's true, but from the perspective of where the market has grown, are there more users on blockchain than before? Absolutely. NFT trading is now about $300 million a month. So not like the peak, which was in 2021-2022, where we traded over a billion a month, but still it's a market that has developed and is also more mature, because you also have to say, 2021-2022 there was a lot of speculation there, so the foundation wasn't there yet. And you see that with stock prices too. It comes back in our opinion, just like other things. I mean, people forget, for example, when Facebook went public at the beginning, the price also collapsed heavily. Google was the same. And after that it found a market and of course the value developed. And not every company will be successful, that's always the case. So in that relation I think the thesis is correct, but it has evolved. Because for example, we imagined that TradFi would go completely into crypto, at least in America and in Asia. We didn't think that, that wasn't in our plan, and that of course changed the market significantly.

So the competition now for Bitcoin and the big tokens is ironically not other tokens but tokenized stocks and equities and commodities. The users who are in the crypto space are now buying tokenized gold, tokenized stocks of different companies, and trading that on the exchanges. So that has changed. And the people who exclusively bought crypto are now of course also buying stocks that are simply tokenized. And you also see the advantage of that, because for example, if something happens on the weekend and there's a war and oil prices change, the normal market can't trade on the weekend. That's a break. Everyone goes home for the day. But in crypto, on the blockchain, they trade 24/7, and then you have developed an actual market that you can trade through, which actually really should be the real market, because if something happens on the weekend it doesn't mean the market stands still. That's the traditional system, and that has changed a lot too. And you see other things, whether you say whether that's gambling or not, things like Polymarket, prediction markets for example, these are new markets that were developed through blockchain which actually weren't possible before because of the costs and so on. So crypto has developed significantly but maybe not quite in the direction one would have thought. And the application of stablecoins is currently I think one of the most important geopolitical issues at all, especially because of the US dollar. And comparatively I think it's a risk for Europe because the euro stablecoin isn't really coming together somehow, because I think because of the central bank, which maybe in terms of focus is perhaps different compared to Asia for example, where that's also a strong focus. And stablecoins, the trading of stablecoins monthly just on Base alone is between 8 to 10 billion dollars a month. So we're talking here about a trade that is larger than some of the larger countries on a blockchain. So it has changed. And BlackRock and Fidelity, Franklin Templeton, all the really big corporations in America and also in Asia are all going in the direction of tokenizable assets. So there are many reasons. The blockchain technology and tokenization is coming. Of course that also means that smaller tokens, just like smaller companies, can also have success but not necessarily to the extent that we had before. And the whole trading becomes more cost-effective and easy. So I talked to Jenny about that at Franklin Templeton. Now products that she sells for $500, she can now sell for $10 or $20 because the costs and the distribution costs are simply much lower. And she can make a profit now selling someone a $10 mutual fund rather than $500, because otherwise if she can't sell it for $500, then she takes losses on it.

Yeah. So it wasn't really the plan for us to become an investment firm, but simply we made those investments back then strategically speaking because we just didn't really have the talent for blockchain technology and we didn't have the experience either. So in 2017, 2018 we were actually beginners in the industry. And how can we build experience and partnerships? We do that through investments. And in 2017, 2018, nobody believed in NFTs. And generally, people were also very critical of blockchain. If you remember what happened in 2018, there was that whole ICO crash. Bitcoin was I think $3,000 or less. Ethereum was under a hundred, I think $80. So that was definitely a crisis period and nobody was interested in crypto back then. So that also means when we made these partnerships and investments, it wasn't about price but rather about we want to work with you and we want to learn the market. And I can remember that was the first NFT conference in Hong Kong. I think that was in March or April 2018. There were maybe 200 people there. And that was the whole industry, everyone in one room. And we were actually some of the only people saying, 'Oh, we would like to invest, oh yes please.' It was very easy. It wasn't like now where there's a lot of competition and many investment firms. And we had really invested because of partnerships. We got very large shares back then relatively speaking. We didn't have a lot of money but relative to these companies we had money, and then we just developed the industry back then through that. But that was the beginning.

And then of course came the thesis that if you want to build in a decentralized system, then it's perhaps better that we build as an ecosystem builder rather than as a classic VC. All of our investments come from our balance sheet. That means we don't have a fund life cycle. We can be long-term and we can also be very strategic. So we're not dependent on a fund life cycle. And one of the big promises of Web 3 is actually that you also distribute the market shares more, so better distributed, more distributed for value, because everyone can then become a stakeholder through tokenization. That means they can have a share of it. It's like if, for example, Uber, all the people who drive for Uber as Uber drivers, that they could then eventually over time earn an ESOP. Would Uber as a company be different if every Uber driver also had shares in Uber, let's say 5% or 10% of the company? Then of course I think the relationship would be completely different. There's much more alignment. The partnerships are just better. The driver doesn't just have more loyalty, but is also then generally more interested in Uber. And even if Uber might cost a bit more or maybe you have to work a bit harder, it's okay. I'm not just working for them. I'm also working for myself. So that's the whole idea of stakeholder ownership and tokenization. Tokenization makes that possible.

And I always thought to myself, why doesn't Apple have an ecosystem fund? I mean that's a company that back then also had a lot of cash, but back then they had even more cash before they did the dividend. Why wouldn't they for example set up one or two percent of their profits as a kind of App Store ecosystem development fund? Because if now many companies were invested in by Apple, then I have a reason to pay 30%. Because I know that if I now pay the tax to Apple, that something comes back into the ecosystem so that it keeps growing. That's exactly the reason why we pay taxes where we live, because we know the street is clean, there are police, there's the justice system, we are safe. That's a reason. So you can't live without taxes. And you have to of course invest into that. But with the profits that these companies make and bring nothing back, of course then the big question is why do you pay 30%. Back then when Apple was growing, we got customers relatively cheaply and easily. That means there's a reason to pay 30%. But now there's absolutely no reason. Apple doesn't provide the same value for 30%. And that's also the big fight that you see between Epic and Apple over this 30%. And I know the EU is fighting against it too. But those are things where if Apple for example now had a fund and said I'm investing one or two percent of whatever cash they have, which is nothing for them, that would make them one of the biggest investors for startups. And secondly, I believe that every company that is invested in by Apple would gladly pay its 30%, because they know that that is more than just a tax that just goes to Apple's profits. And that is just the big topic of stakeholder alignment, which tokenization makes possible.

I mean we have about over 600 companies that we've invested in now. We have hundreds of employees all over the world. I think we had our last audit out, but revenue was around 200 million, 300 million US, EBITDA positive. And we're sitting on billions of dollars in crypto holdings. So we are let's say a medium-sized company in the crypto space. We're relatively big for what we are because it's not an exchange. So I think in terms of what we do, we're probably the biggest, but of course as a company we're still relatively young in comparison.

Yeah. So I can't talk too much about that because it's obviously public. But that we want to be a public company, that's definitely clear. And how we do that, of course, we had announced a merger back then. But of course there are, we also see, so that's a NASDAQ merger, it's called Current Group. But of course we also have other partnerships because of that. So I can't really go into detail about that unfortunately. But in any case the plan is for us to go public, and America makes the most sense as a market because in the end it's the largest capital market and also when it comes to technology they are probably also generally more forward-thinking than elsewhere.

Yeah. So generally for founders, since a lot of it is early stage, the founder is incredibly important. There are of course skills that you need. So I mean nowadays it's very important that the founder can really apply AI well, especially agentic AI, because that's not just a cost thing but also efficiency. But if we're talking about talent in general for technical things, all things being equal, I think what's important for us is that the founder has, as we describe it, I don't know how you describe this in German, but like a chip on the shoulder. Something that is very important to them, that it's not about the money. Because if it's only about the money then there are many other ways to make good money. And of course, if you're short-term, maybe you can work with someone like that. But if you think long-term, because many of these companies, it takes years for them to develop and to reach a certain size, it's important to understand why they're doing it. What is important to them in terms of their background or the reason why, or is it about something that is philosophical or moral or personal, something that is so important to them that they absolutely have to succeed. Because things will always change and you just have to understand what the big reason is for them. And that's not always easy to find, but that's one of the things that are very important for founders in the early stage. Later when the company is bigger and obviously has revenue and a business model, then it's maybe less important because then you also have management and so on. But for the early stage it's definitely very, very important.

In terms of where we focus, so that obviously includes blockchain and Web 3. But the difference is that we now generally, the thesis has evolved, that we believe AI agents are the future of the web. So we describe the next phase of the web, Web 4, as the agentic web. The web where there are more AI agents than humans. And it sounds logical, in our view, simply because every human will have at least three to five, or ten, or twenty, or if you're in San Francisco they say two hundred, three hundred AI agents. In any case there will be more AI agents than humans. And the AI agent can do many things that humans can't. It can do a thousand transactions in an hour or maybe even in a minute. It doesn't sleep. It can trade things during that time, can gather information. And you can already see it now with just normal AI. When was the last time you asked things through Gemini or ChatGPT and it says, 'Here is a website and here is information.' The information was enough. People don't click through to the website anymore. That has already changed. Now if there's an AI agent that wants to do something for you, and that will then be transacted, it needs a wallet or a bank account so that it can pay. How can it do that? Through blockchain. So the same technology that made it possible for people in the Philippines who for example couldn't have a bank account but then could be financially included, can now do that for an AI agent but on scale. So that's the machine banking system.