CZ's Miami speech: Cryptocurrency will become the native currency of AI agents.

Compiled by:Wu Shuo Blockchain

This article is from a fireside chat by Binance founder CZ at Consensus Miami 2026. CZ admitted that he had deliberately avoided the US in the past few years, but observed a 180-degree turn in US crypto policy over the past year and a half, and that the US is now leading the world in terms of legal clarity (such as the CLARITY Act) and talent repatriation. He pointed out that although the US has top talent and a financial market, it still lacks the world's best crypto liquidity, and the BNB ecosystem will be committed to filling this gap and promoting the development of the Binance exchange and BNB Layer 2 (opBNB) in the US market.

In terms of technology outlook, CZ believes that cryptocurrencies will become an inevitable choice for AI agents.

Crypto Policy Assessment: US Regulatory Shift Leads the World

Michael Lau: You haven't spent much time in the US recently. This is your first major public exchange with us recently. What does coming here mean to you?

CZ: I want to be very frank, for the past few years, I've been trying to avoid the US. So I've been trying to keep a low profile, out of sight, out of mind. But I think it's clear that US policy toward cryptocurrencies has changed in the last year and a half or so. Now we have to make up for that previous absence.

Therefore, I think we previously lacked contact with the US community, lacked communication with US developers, regulators, policymakers, etc.

So I think we lacked contact with the US community, lacked communication with US developers, regulators, policymakers, etc.

Furthermore, there are quite a few misunderstandings about me, BNB Chain, and our ecosystem. So I'm here to spend more time communicating with more of our community members and hopefully resolve these issues. Michael Lau: You travel the world speaking with all these people who have significant influence and influence in policymaking and regulation. I'd like to know, could you give us an assessment of what's happening in the US? How does it compare to what's happening elsewhere in the world? What are your thoughts on the current situation? CZ: I think the US has a very strong document, the US Constitution. So the US can make 180-degree turns on certain policies and correct itself very quickly. The US has fundamental advantages: talent pool, financial markets, and a traditional regulatory framework, and the current leadership is very forward-thinking. So things can change very quickly. I believe the US is currently leading the world in crypto policy. Most other countries aren't moving as fast, nor do they have the confidence to quickly implement crypto regulations. Clarity of regulations is being discussed in various new bills. While some minor issues remain, the US is moving very rapidly. I believe many US crypto professionals are now returning. Many left the US before; they went to Abu Dhabi, Hong Kong, Singapore—many developers left without intending to return. But now we're seeing developers coming back. Currently, the US is slightly lacking in optimal liquidity. The best liquidity in the crypto space is outside the US. Therefore, the crypto market is one of the very few markets where the US cannot obtain the best prices. Hopefully, we can help address this issue. I believe Binance has the best liquidity in this market within our ecosystem. We very much hope to provide that liquidity in some way. Whether it's bringing Binance back, revitalizing Binance US, or providing American consumers with the best liquidity and the best prices in the world in some other way. BNB Ecosystem Value Reassessment: Filling the Liquidity Gap and Unlocking Institutional Funding Potential Michael Lau: Speaking of liquidity, I'd also like to briefly mention BNB. Are you also considering liquidity in the DeFi direction? CZ: BNB Chain has also tried to move away from the US market in the past few years, so their business presence here is insufficient. I think they've made some changes since last year. They now have a developer home in New York, and I believe they also have a small presence in San Francisco. They're also hosting more and more events there. YZi Labs is also investing more in the US now, something we've been trying to avoid doing for the past few years. So now everything is on track. I don't think BNB Chain has gotten much exposure in the US. I think other Layer 1 blockchains have done more marketing, community building, and establishing developer homes in the US. I think this will change, and it's already starting to change. I also hope to see more of these initiatives. Regarding the BNB token, US institutions only recently gained access to BNB. Therefore, BNB ETFs aren't as widespread or developed as ETFs for other tokens. BNB is somewhat lagging behind in this regard. However, this lack of access for institutions is actually an opportunity for BNB investors. Institutional funds are generally beneficial for the token. So there's still a lot of room for growth in the US market. Michael Lau: Let's talk about BNB a little more. As you just mentioned, the focus in the US over the past few years hasn't been really on promotion or education. We have a lot of people here doing on-chain development. If I were to give a presentation, considering that this information and knowledge may not have been well disseminated to the US recently, why would they consider developing on BNB Chain? CZ: BNB Chain is actually a very large, comprehensive, and complete ecosystem. It's actually the most active blockchain network. It actually contains three blockchains: BNB Smart Chain, opBNB as Layer 2, and a decentralized storage blockchain, Greenfield. Most importantly, there are many different ecosystem participants within BNB Chain. Here you have PancakeSwap, one of the most active decentralized exchanges, and Aster, a perpetual contract decentralized exchange that's currently ranked second in trading volume. While it's still a bit smaller than Hyperliquid, it's second and growing rapidly. You also have lending protocols like Lista and Venus. These are very popular in Asia, Africa, and other places, but not so much in the US; I don't think the US community has really gotten involved in these areas. Then, in the same ecosystem, you also have Binance, Trust Wallet, and CoinMarketCap. So it's a fairly complete and very active ecosystem, just not widely used in the US, which effectively excludes US consumers. So we hope to address this issue. Industry Trend Assessment: AI Agents Will Reshape Payment and Transaction Models with Cryptocurrencies Michael Lau: Is Real-World Asset Tokenization (RWA) Undervalued or Overvalued? CZ: Actually, I now think it's undervalued. A year ago I thought it was overvalued, but now my opinion has changed. I think it's real; RWA is a tangible thing. Michael Lau: Okay. What about institutional adoption of cryptocurrencies? CZ: It's just getting started. We're just starting to see a trend. Michael Lau: What about our industry's obsession with conferences? CZ: I think it's great. I mean, for us, there are so many conferences that it feels repetitive. But I think it's important that the people are slightly different each time, and the locations are different too. It's also a great way to get builders together for short, quick meetings because everyone is gathered here at the same time, and you can schedule a lot of quick meetings. Otherwise, if you had to schedule meetings with them separately, you would have to go looking for them, and each time it would take several hours. So it's a great way for people to connect. I think the conferences are quite good; Consensus has done a fantastic job over the years. Michael Lau: I think you made a comment earlier this year in Davos that the native currency of AI agents will be cryptocurrency. And you also believed that agents will make millions of times more payments than humans. Could you elaborate on why you think this is an inevitable future? And why cryptocurrency, rather than more traditional payment channels? CZ: Let's review that briefly. In my adult life, there have truly been three fundamental technologies: the internet, blockchain, and AI. All three are borderless, global, right? So you can send a message to anyone, anywhere in the world. You can do it at 3 a.m. today. But you can't send money at 3 a.m. on a weekend. Money is nationally controlled. But now we have the technology to make currency borderless. It should be a highly aggregated, low-cost, instant-operating global marketplace. Generally speaking, it's permissionless. AI is also a borderless technology. Similarly, AI is more centralized than cryptocurrency. So only a few AI companies will be drawn into geopolitics for reasons like national security. I'm a tech person, so I don't touch those things. But fundamentally, your agent can interact with agents in another part of the world. For example, if your agent interacts with someone in Pakistan—perhaps a programmer, a designer, someone who works for you, or vice versa—you need to make a transaction, so that transaction could be between your agent and his agent. Or, a person from Singapore wants to attend a conference in Miami. Their agent can talk to an Expedia agent to book flights and hotels. The agent will take the time to search for you. For example, the hotels here are expensive and fully booked. Which ones are within budget, which ones have availability—all of this should be done by the agent. Then, when the person in Singapore wants to pay, a credit card works, but the agent doesn't know how to swipe it. Credit cards don't have an API, so the most native thing the agent uses is obviously blockchain. If Expedia actually did accept cryptocurrency for a while—I don't know if they still do now. If they do accept your cryptocurrency, then direct transfers by the agent are very simple and very fast. It's the most natural technology. This is just how we use it at a very basic level. I believe there will be a huge demand for smart agent payments in the future. And smart agents will be handling more microtransactions. It's like, I don't want to spend $200 to subscribe to a newspaper, but sometimes I'd rather spend $1 to read an article. Then I want my smart agent to pay for it. For that $1, I don't want to manually enter my credit card information and all that stuff. Smart agents will be making transactions very frequently. Today, if you open the Binance app, you see a bunch of charts, a bunch of numbers, and then you have to choose a chart, zoom in, find the price, enter the price and quantity on your phone, and then click the red or green button. It's very counterintuitive. What you should do is open your phone and say, “Show me my portfolio. What happened in the last 24 hours? What changes have occurred in my portfolio? What news is there? Has the Clarity Act passed? Is the war over?” Then say, “Okay, I want to exchange 10% of my stablecoins for BNB.” If you have a large position, this should be done in the background by an agent. The technology that makes this work today already exists; it’s just that no one has built it yet. So, interacting with an agent is a more natural way of human interaction. If you expand on this, I think all quantitative firms are already using AI to some extent for training, data analysis, etc. In the future, it should be agents doing the trading. All exchanges need to be agent-ready. I think AI is a foundational technology that can change many different parts of our world, but AI still needs money. So blockchain won’t disappear. AI won't suddenly make us no longer need money just because we have intelligent agents. Money is a fundamental concept in economics; it drives everything. So blockchain won't disappear. I see many blockchain developers moving into AI. You can work on AI. But if you look at Nvidia, for years graphics cards weren't the important thing; CPUs were. So it took them 40 years to get to where they are today. Now AI is here, and they've become the most valuable company in the world. So I think for blockchain developers, money won't disappear. For me, YZi Labs invests 70% of our money in blockchain. Then 20% in AI, and maybe 10% in biotechnology.

Industry Layout and Personal Planning: BNB Chain Aims to Become an Intelligent Agent Currency, CZ Focuses on Education and Founder Mentoring

Michael Lau: Regarding the future of intelligent agent business you described, how will the BNB ecosystem respond? Is this the direction they are currently focusing their resources on? In what specific form will it take?

CZ: I think this is just the beginning of AI. All blockchains need to be prepared for AI. You need to support intelligent agent payments, you need to have open capabilities or skills similar to Claude, it needs to support everything related to AI. This field is very new and will continue to evolve. For example, those AI capabilities that can control computers have only been around for about two months, and now they have already been copied by top players.

This is the first time an intelligent agent is not just chatting with you, but performing operations on your computer, actually doing things for you. I think this will develop very quickly. We are still far from the final form. If the whole process is a 24-hour journey, it's probably only a millionth of a second that has passed. So we will see more innovation, and now is the time to build the infrastructure for AI. There are many people building large language models and other related things, which is great. For us, I think the goal of BNB Chain is simple: it should become the currency of intelligent agents. Michael Lau: Who is CZ today? What is your vision? What are your goals? How should we view you now, and how will your new chapter unfold? CZ: I think people come into this world hoping to make a positive contribution. If you have the ability, you should do it. When we leave, we can look back and say, "Look, we made some positive contributions." You should take care of yourself first, and then make positive contributions. If your abilities are limited, make small contributions; if your abilities are strong, make larger positive contributions. That's my mindset. I believe that any animal is genetically selected in this way. If any animal doesn't want to help its own kind, the species' competitiveness will decline, and it will eventually be eliminated by evolution. So since we still exist and are the dominant species on this planet, our genes dictate that we want to help others. When we do this, we feel very happy; it's an intrinsic reward. When I founded Binance, our goal was actually to become a top-ten exchange within 3 years, but we became number one within 5 months. We were thinking, what happened? We didn't plan for it that way; we just wanted to provide the best products, the best servers, the best security, and the best user experience. We didn't know it would happen so quickly. Once we got there, and of course, we've been leading for almost nine years now, that's what I've always strived for. For me, that was never my initial goal. I never said I wanted to be the best, or to climb the Forbes rich list; in fact, I'd rather be on that list. So my goal is really just to contribute as much as possible. As long as you can tell yourself you've done your best, that's enough. For me, I don't care much about fame, power, or even money. People need enough money, but I already have enough. Money beyond that limit doesn't help me much. So my goal is that when I'm old and look back, I can say, "Look, I did my best." That's my attitude. With that mindset, I just want to continue striving to make a positive contribution. I don't feel like I have the energy to run another startup or lead another company. I'm kind of like someone who only knows one thing, and I've accepted that; I've retired from the game. So I want to help other founders, other entrepreneurs build their businesses. We invest in them, and if they're willing, I can mentor and guide them. If they're in the crypto space, I can give them advice on token economics, pitfalls to avoid, how the industry works, and what to expect when launching a token or project. I can also advise ordinary entrepreneurs, guiding them on how to build a product, how to build a team, etc. I always tell them that my methods are almost certainly not the best, but they are one approach that people can consider. I want to help other founders succeed; that's my indirect contribution to the world. I allocate my time across four different areas: helping founders at YZi Labs, supporting ecosystem builders on BNB Chain, and spending time on Giggle Academy, a very interesting project. Giggle Academy is providing free education to 260,000 children today, and the project is only a year and a half old. Every time I talk to them on the phone, I feel incredibly happy. That's the intrinsic reward you get when you're working on something truly meaningful. Then, the fourth thing I do is advise governments in different countries on crypto regulatory policies. Only a handful of governments, maybe a dozen or so, need my advice, and I'm happy to provide it. These include countries like the UAE, Pakistan, Thailand, and many African countries. So that's how I allocate my time. I'm much more relaxed now, and my work isn't as hectic as when I was running Binance. Back then, I worked about 20 hours a day, constantly attending 30 to 40 meetings. Now I work about 10 hours a day, attending some shorter meetings, which is much easier. Life is pretty good now.