Web3 Titan Yat Siu: How Digital Property Rights Will Unleash AI Agents

Yeah, I mean, I guess it's a bit of a long story, so I'll try to keep it short. I grew up in Vienna in the 70s and 80s. My mom was a musician, and there was a time where she was working at the eastern side of Berlin, at the Komische Oper, and that's also the reason why I speak German, because I basically grew up over there. And sometimes I would visit my mom, and she was at the eastern side of Berlin, and I would cross the border. And it was really a strange feeling, although even back then when you were in Vienna, you would actually cross the border into Hungary, for instance, you would go to Budapest. Actually, a similar experience where they would not just look at you, but when you basically cross the border, if you're in a car, they would look underneath the car. You have people with machine guns. All the stuff that you see in those spy movies has actually really happened. But of course, back then, things were kind of normal. This was the norm. This was not something that we thought was weird, except when you go to the other side, you would see the market, and you'd basically have cans spread out literally one meter apart from each other. And while I didn't fully comprehend what it really meant, I knew that this was not the world I wanted to be in. And I think this understanding of what true extreme forms of socialism really means. And again, often this is where the West conflates things, because what they've experienced as communism is very different from what Chinese people have experienced in their version of communism, which is actually much more capitalist, perhaps more capitalist than most of the world. That was sort of shaping the worldview, and what does it mean to be in a world where you don't have freedom? What does it mean to be in a world where you don't have property or the rights to property? And I think today, fast forward a little bit, having worked in Europe, then having worked in the U.S., and then coming to Hong Kong, my father's from Hong Kong, so I had permanent residency so I could work here. I didn't realize, it was more of a happenstance. This was in the 90s. I set up one of Hong Kong's first ISPs, Internet Service Providers. I sort of saw different extreme forms of capitalism as well. In other words, going from Europe, which is a socially democratic place, to America, which is an extremely capitalist place, going to Hong Kong, which in the 90s and 2000s was hyper-capitalist. I often quote it as Milton Friedman's sort of wet dream in terms of how you have a system that is entirely based on just capitalist values, but we didn't even have social security. So when I first came to Hong Kong, there was no MPF. MPF was established, I think, maybe in the late 90s or early 2000s, which meant people didn't even have a social safety net. I mean, that's crazy to think today. So you get the different extremes, and in a way, you still see the state in Hong Kong. You still see 70-year-old senior citizens carrying out the trash, which is also not okay. But then you have the other extreme where basically people are sort of against anyone who has wealth, which is becoming more and more of an issue in Europe. So it's helped me shape the thinking. Now, when I started to encounter what happened in blockchain and Web3, we were already in technology a fair bit. We came to realize that this was a way in which we would actually own our digital property. So our thesis around digital property rights started to develop. And it's a very simple thesis, which is to say, property rights gave us wealth in the physical world. And where is most of the wealth being generated today? It's actually digital. So who owns our digital property? It's all the big platforms. Facebook, Google, Apple. We're renting them. And we don't even think we're renting it. Every time we post something on Instagram, every time we're sharing something on TikTok, we're actually enriching their algorithm. We're making them more powerful. And we get likes and we get social credit, but we don't make any money from this. And today, it's even more pertinent because all of that data we're giving is essentially fueling the AI that is taking over the world and making these companies ever more powerful. And yet we as users don't get anything for it. So in a way, we're constantly delivering free data to the platforms, and we the users get nothing for it. And so that was the thinking around, how can we change this? How can we truly own our digital property? Blockchain technology was a solve.

The biggest pattern is, of course, this is a recognized pattern. Everything's becoming more digital. That's clear. This has been going on for the last 20 years. However, the big movement right now, when you talk about digital, when you talk about what we used to call the metaverse, or maybe still call the metaverse, when you talk about blockchain, tokens, the whole cryptographic technology, our pattern that we're seeing is that everything we've built over the last 7, 8, 9, 10 years really wasn't meant for humans. It was meant for machines. And all the technology, including Bitcoin, was really the proof of concept for the machine economy. So to this date, it remains difficult for a user to use a MetaMask wallet. They're used to classic fiat. And if you think about the legal system and the whole technology of blockchain and smart contracts, what's this idea that code is law? But humans don't operate with code is law. If someone steals your Bitcoin, you still call the police. You still want the judiciary to help you. Because under the human construct, it doesn't work this way. But under a machine construct, it does. And so all those codes and bits and everything that you see, the smart contracts, are much more understandable and provable in the machine economy. And I think a lot of people miss the point because they try to divide blockchain and tokens separate from AI, but they're so intertwined. Now, so I sort of conducted a survey of people in the room, and most of them do not use crypto, and that's fine. But I asked them, how many people of you think you'll use an AI agent in the future? Almost everyone raised their hand. And then I asked them, how many think it'll be between three to five? Maybe two-thirds. Five to ten? Maybe the remainder? The point being is that somewhere between five to ten agents is the minimum number, and we think probably more. I have over 40 agents today. So if that's true, every person who's online today, which is around six billion plus, is going to have anywhere from 30 to 60, maybe 100 billion agents. Where do you think they're going to live? Are they going to live in a centralized platform? We hope not. And the reason why is because the next layer is going to be agentic commerce, in particular agent-to-agent commerce. We get so used to this idea, to talk about patterns, around removing the middleman. And removing the middleman is what we've done with the internet. We've removed a classic physical middleman for the digital middleman because it's less expensive. What's even less expensive? Direct. If I could actually do business with you directly. But the scaling and the discovery doesn't work for humans because we don't think in that kind of speed and we're unable to. But an agent doesn't have that problem. So meaning we're going to disintermediate the classic platforms because agents can do business with agents directly. And essentially, for us, the equivalent of light speed. And they're going to do it 24 by 7. When the whole Iran war started, basically, in classic Trump fashion, it happened just before the weekend. So markets couldn't react. But the crypto markets were reacting. People were essentially able to trade and buy digital commodities because the market was 24-7.

Absolutely disintermediate. And also, we can stop thinking about so many ways, like advertising. Who are you advertising to? Not a human. You don't need to advertise to a human anymore. You need to advertise to an algorithm because your AI agent is going to present your option A or B or C, not the whole spectrum of 100,000 options. And so advertising, marketing, sales, communication, everything is going to be evolving very quickly. But when it comes to transactions itself, again, looking at my business, so painful — PayPal, Stripe, the percentages that they take for any transactions, it's just not 2.5%. We're talking about exchange rates. If your account is, for example, in Hong Kong dollars and you're transacting in the U.S., you end up with 5% to 8% transaction fees.

I mean, you talk about robbery, I mean, this is like mafia.

Yes. So what I'm saying is that there is also that, let's call traditional mafia, right? That does not really want... They're sitting there as godfathers and they basically control the flow of currency, the flow of business, so to say. I mean, they're not going to go down without a fight.

So first of all, and I do want to sort of make one point around building agents. I actually think that everyone will have an agent and you will no longer need to be an AI expert to build agents. And what I mean by that is when you think about what's now possible for an agent to program stuff, the issue that people have is not, can I tell my agent to make something for me? Because actually you can do that. And it's no longer just a case of, hey, I'm using Cursor or using Replit or something like that. Literally the agent, in Animoca Brands, I'm just saying, hey, integrate basically Duffel API or give me a flight service. They can literally build a customized dashboard for you that basically helps you manage your travel service. Literally like your mini travel agent. But the issue most people have is they don't know how to ask the question. Because we've been so trained to search for stuff. One of the geniuses of ChatGPT when it came out was it wasn't just that it was really good AI at the time, LLM. It was just much better search. And it came in an interface that you knew, I ask it a question and I get a better answer. What we're not used to is giving it a task and saying, come back to me. So what agents are doing now is I'm probably as productive as ever and I'm probably busier now than before because I have agents coming to me asking for decisions, which is different. When you have an interface where you're asking the question, it's not about decision. You know what you want and you're asking the question. Here, they're basically like, I need to survey this research on some market stuff or some product or some marketing insight and they literally come back to you and give you one insight and then six hours later, they come back, we found something new, and then three hours later, they might say, did you look at this? So basically, it's like an employee that's giving you feedback and input. But you need to know how to ask the question. You need to say, hey, I'm looking for this. I'm interested in where the market is and you have to give it context. You have to say because I want market share or because I want to grow in China or in Hong Kong or whatever it is. You have to talk to it like an employee and people aren't quite used to that at the moment. That's part of it. But once you know how to ask the question, you're basically off to the races. And we're so good at this point as a technology overall that you just have to prompt it. The image I often have around startups when you look at entrepreneurs, like the U.S. is like the hero cult. If you are a successful entrepreneur, front cover, here's the guy. It's about the individuals, individualism. He changed the world. Even though he had a whole team behind him, it's about the guy. It's always a story. Everybody knows him by name. Whereas in China, you do have that a little bit, but often when you hear them speak, they thank their country, they thank their people, they thank their team. In fact, in Asia, it's much more of a collective spirit. And I think that translates down as well into the rest of the culture to feel like, well, I'm building something, then I have some obligation to return something to the nation I built it in. There's a sense of that obligation. So I think there's a philosophical difference here as well, which I think also means how you build technology and AI and tooling will also come from that mindset. Also, I think in China, it's hyper-competitive, down to the core. I mean, you know this, of course, incredibly well. People totally misunderstand China because they look at China and think it's Russia. State-owned, state-controlled... State-directed, state-invested companies that are competing amongst each other. It's nonsense. They have no idea what they're talking about because they still look at China and they look at the CCP and look at the communist world and in their mind, communism is Russia, which is not really communist either, frankly speaking. But the point is just that they have this whole wrong perspective around what that means. China is more capitalist than probably most of the world. The competition is not only within businesses, but it's also within industries themselves, within states, cities. They compete with each other for prominence. Which is how they provide funding and grants and support. So it's like a higher ecosystem that really is, I think, unrivaled in that sense. And that's actually fueling it. Where does the money come from? It doesn't just come from China top-down saying, I like this, let's put money into that. It comes from basically local state enterprises competing with each other for essentially the ability to climb the ranks as well. So there's a lot more competition there. And then Europe has none of that. So unfortunately, they have neither U.S. or Chinese competitive edge, which means, unfortunately, Europe is really in trouble, I think. But that's a different topic. So between that, that sort of creates this environment. But it also means that the risk of that danger, like you can be more dangerous in America, you can't do that in China. And I think that's the edge that America still has. You can create that disruptive technology that could challenge government, which is absolutely not possible in China and shouldn't be. For China is all about harmony and stability. U.S., harmony and stability, what's that? They live for that disruption. So there's something unique about America. And so I think the world needs that because if we didn't have that, then all of the crazy innovations that have come out of America that have influenced the world and, of course, caused the reaction in China to say, we need that too. And it's not the government. It's entrepreneurs in China saying, we need that. And they're all building and competing and they're basically advancing that. Electric self-driving cars came from America and China just made them so much better, as an example. So there's all these elements around there that I think sort of benefit each other. But of course, you have two nations, incredibly powerful. There's this clashing element, but they're also so intertwined.

Well, I mean, again, for us, we hope, obviously, we'll continue to grow. It has been a rocky road. Our industry has been going up and down. We expect to be a publicly listed company at that point, but let's see where it goes. At that point, we expect RWAs to become multi-trillion dollar asset base. We expect billions of people to have been onboarded into Web3, but not directly, but through their agents. We expect essentially agents to be commonplace. Because we're talking four years from now, but it's progressing super fast. And we certainly hope that our agents or agents in our portfolio are going to be amongst the most impactful agents that are used by everyone. And of course, hopefully, today we have many unicorns in our portfolio, but we hope to have many, many more companies that are building and impacting. And finally, I would say, in a world where we have digital property rights, I hope that we will have onboarded billions of people to have digital ownership. And from that, generate value, not just the fact that they have their data access, that agents are making money online, but they have a stake in what they see. And I think people talk about we're hitting the one-person billion-dollar company. Well, that's one way to look at it. And I think we're already seeing the evolution of that with AI. But actually, what I'm much more interested in, what I think will happen, maybe not in 2030, but in the years to come, is that we're going to have billion-dollar companies owned by millions of users. So meaning from a stake perspective. And that's what excites us. If you think about Bitcoin and Ethereum as an example, it's owned by millions of people who have helped it grow and have had a share in that stake. But still, it's the blockchain community. It's relatively small compared to the rest of the world. And this experimentation we did with DAOs, decentralized autonomous organizations, where you can democratically sort of decide what goes with the protocol. Well, humans were not super efficient on this, but maybe agents can change this. The point is, now we have a framework that humans tested, but agents can execute around a system where we could co-own a network, maybe worth billions, but owned by tens of millions or hundreds of millions of people who share in its value and contribute to its value that could be powered by AI. We expect those to really sort of evolve. Probably by 2030, we'll have some very good examples. We needed blockchain technology to show what we could do with early forms of DAO. We needed agents to get to this point to converge and we're going to have new kinds of companies. I think we are no longer going to have just companies. We're going to have basically these decentralized autonomous organizations that are governed essentially on chain.