Binance founder CZ talks about FTX, Hyperliquid, and the 1011 incident

Original Title: CZ Binance Founder: This Crypto Bull Run Could Make You Rich

Compiled by: When Shift Happens

Editor: Wu Says Blockchain

Binance founder Zhao Changpeng (CZ) was interviewed on September 25 by the When Shift Happens Podcast, reflecting on his journey from an immigrant family in Canada to forming an early team and founding Binance. He believes that Binance's early rapid growth was due to the team's execution, user protection, and grasp of industry changes, while the current crypto market is still far from saturated. CZ also discussed the performance of projects listed on Binance, ordinary investors participating in early projects, the digital "4" meme, the FTX collapse, experiences with the U.S. judiciary and presidential pardons, and the different handling of the Binance hack in 2019 and the Bybit hack. In the latter half of the interview, he responded to competition between Hyperliquid and Binance, HYPE not yet being listed on Binance, the controversy over the market crash in October 2025, and shared reasons for investing more energy in crypto regulation, YZi Labs, and Giggle Academy, as well as his views on AI education, nurturing the next generation, and the long-term development of crypto.

The guest's statements do not represent Wu Says' views and do not constitute any investment advice. Please strictly follow local laws and regulations. Audio transcription and translation were completed by GPT and may contain errors.

From an Immigrant Family in Canada to Crypto Guy

Kevin: You just arrived and are leaving again? This coming and going is basically your current lifestyle?

Zhao Changpeng (CZ): Pretty much, and I'm a bit tired right now. When I go somewhere, I usually only stay for one or two nights and then leave because there are always other things to do.

Kevin: How do you handle this pace?

Zhao Changpeng (CZ): I don't know. But it's actually much easier now than before; I have more time to arrange things for myself.

Kevin: How do you decide where to go and what events to attend?

Zhao Changpeng (CZ): It's not really my decision; I usually let the team choose. My team is very good at handling these matters, especially for events, who to talk to, which podcasts to go on, it's basically all arranged by them.

Kevin: If I told you that the biggest change in your life hasn't happened yet, what do you think that change will be?

Zhao Changpeng (CZ): I believe the biggest change hasn't happened yet; I just don't know what it will specifically be. I think there will always be more opportunities in the future than in the past, and the speed of change is increasing, as well as the magnitude. Before the internet, it was very difficult for the whole world to change simultaneously; now information flows more freely, we have a network about "money," which is blockchain, and now AI has emerged. So I believe there will be many huge changes in the future; I just don't know what they will specifically be. AI will bring a lot of efficiency improvements, and those improvements will further accelerate these changes.

Kevin: Let's start from the beginning. Who are you?

Zhao Changpeng (CZ): I'm just a crypto guy, or you could say a crypto sales guy. President Trump once called me "the crypto guy." I am a very firm supporter of blockchain technology, and I believe that so far, my biggest contributions have mainly been in this field, and most of my future energy will still be focused here. I also go to many different countries now to help them discuss issues like crypto regulatory frameworks, so people usually see me as a promoter of crypto.

Kevin: Many people we've interviewed have a common point; something happened in their lives, maybe a challenge or some kind of trauma, that ultimately gave them a drive to prove themselves. What happened in your life that can explain why you are listed among the world's wealthiest people before the age of 50?

Zhao Changpeng (CZ): First of all, I don't think those rankings are accurate. I might be in the top 100 or maybe the top 500; I don't know. There are many very wealthy people in the world who aren't on those lists, and I think the external valuation of my net worth is seriously inflated.

Kevin: So where does your drive come from?

Zhao Changpeng (CZ): I can't recall any particularly severe trauma from my childhood, but my family was indeed quite poor when I was young. My whole family immigrated to Canada, and my parents' income was close to minimum wage at the time. In contrast, some of my friends came to Canada through investment immigration from Taiwan, Hong Kong, or other places, and their families were relatively well-off, so there was indeed some comparison and pressure when I was young. But overall, I didn't have any particularly terrible childhood experiences. I have many good friends from Taiwan, and they and their families have been very kind to me; it's just that there was an objective economic comparison.

I actually feel very lucky because I stand on my parents' shoulders. My parents immigrated to Canada for my sister and me to have a better future, so I basically count as a second-generation immigrant. People often say that second-generation immigrants see how hard the first generation has it. My parents didn't speak English very well at the time, and first-generation immigrants often have to do jobs that pay close to minimum wage and are very hard. The second generation usually gets motivated to keep moving forward after seeing their parents go through this, and I think that may have given me some drive. Typically, by the third or fourth generation, that drive may start to weaken.

Kevin: You just mentioned your sister; what does she do now?

Zhao Changpeng (CZ): She also has a technical background; she used to work at a startup in Tokyo, then went to Morgan Stanley, and became a Managing Director at a young age. There are probably only a few hundred Managing Directors globally at Morgan Stanley, but the company has a lot of employees, so her career development has actually been very successful. However, she developed her career in a large company and retired around the age of 42. After having a child, she experienced a period of postpartum depression, then left her job, and about a year later, she fully recovered. Now she also helps and consults with people experiencing depression.

Kevin: You also have children. You just said that by the third or fourth generation, there may gradually be a loss of the drive seen in the previous generation. How do you avoid your children having this problem?

Zhao Changpeng (CZ): I don't know; I haven't found a truly effective answer yet. And my children don't have children yet, so this question can be pushed a bit further down the line. I occasionally think about this; I believe education may be part of the answer, but it's indeed very difficult.

The Starting Point of Binance: Team, Execution, and Entrepreneurial Culture

Kevin: There's a famous video of Jack Ma in 1999 at home talking about Alibaba's future to 17 co-founders. News reports always have a startup story, but the stories in the news often don't match the reality. Take me back to the first day of Binance's founding; if I were in that room, what would I see or hear?

Zhao Changpeng (CZ): The way Binance started was a bit different. At that time, I already had a team of about 15 to 20 people, and we were running a company called Bijie Tech, mainly selling exchange systems to other trading platforms. About 80% to 90% of our clients were not crypto exchanges but traditional trading venues, like markets for trading stamps, jade, etc., but they also used order book systems.

I had been leading this team for two years, and the company's business was relatively stable. Then around May 2017, I called everyone into a meeting room and said it might be a good time to start a crypto-to-crypto exchange. I told everyone that if we don't do this now, this idea will keep coming back to our minds again and again. If I don't do this in my life now, I will probably do it again at some point in the future.

I believed we had many of the right conditions; we had products, systems, and a team, but we were just missing a few parts, like a marketing team and a customer service team. Bijie Tech was a B2B business, so we had sales and account managers, but we didn't have a customer service system aimed at a large number of ordinary users.

Then I called everyone into the meeting room, and everyone said, "Okay, let's do it." After that, we discussed domain names, and there were two options; in the end, we chose Binance. That was the moment Binance truly began. The meeting room was actually quite small, probably smaller than this room we are in now.

Kevin: If I were a fly on the wall in that meeting room, what would I see that other crypto exchanges' offices didn't have? What allowed you to grow from zero to the world's largest exchange in about six months?

Zhao Changpeng (CZ): I don't know what other exchanges are like, so it's hard to compare. We didn't have anything particularly magical; we just had a very hardworking and down-to-earth team. Very few people in the team had a strong sense of self-awareness; everyone was very humble, and the execution was very strong. We were also very synchronized with each other; everyone in the team trusted me and was basically willing to follow me.

There might also be a bit of Asian cultural influence here. In Western culture, many times before a decision is made, everyone discusses it, and each person needs to express their opinion to build consensus. But in Asian culture, once everyone chooses to follow a leader, they sometimes follow very firmly, to the point where it’s hard to get them to contradict their own ideas. If the leader's direction is correct, this culture can lead to very high execution efficiency; but if the leader's direction is wrong, it can often be a disaster. Fortunately, we were quite lucky at that time and generally chose the right direction.

Kevin: How do you define "hard work"?

Zhao Changpeng (CZ): "Hard work" may mean different things to different people. I think what's more important than working hours is whether you can produce results and do so efficiently. Many people work long hours but don't actually create anything substantial, so what's really important is whether you can produce high-quality products that users genuinely like. Of course, generally speaking, very passionate people will also invest a lot of time. You shouldn't just measure how many hours a person works; you should look at how much they love what they do and how high-quality results they can produce in a short time. But to achieve this, you usually need very passionate people, and they do tend to work long hours.

Kevin: But if you really love it, you might not even feel like you're "working," right? That's also why some people can invest more time than others.

Zhao Changpeng (CZ): That's basically it. Especially in the early days of Binance, even to this day, I think many people actually don't count how many hours they've worked. If something needs to be done, sometimes after putting the kids to bed at 9 PM, you just continue to finish it, maybe completing it before midnight or even working until 1 AM. That's how things get pushed forward.

Kevin: What is the biggest misunderstanding people have about Binance?

Zhao Changpeng (CZ): There are actually many misunderstandings. Some people think Binance is a Chinese company, but it is not. We do have many Chinese employees. Some people also believe that centralized exchanges are bad for the crypto industry, even evil, but I don't think so. In fact, we have always been promoting and supporting decentralization. However, today, most crypto users still prefer to use centralized exchanges; they are used to using email, passwords, and customer service, rather than managing wallet addresses themselves. But I believe that in the future, it will ultimately become more decentralized, so we will also actively invest in decentralized ecosystems.

Most projects will fail, should retail investors participate earlier?

Kevin: Why do so many projects launched on Binance end up disappointing investors?

Zhao Changpeng (CZ): I think if you look at any market or industry, by the number of projects, most projects will ultimately fail. How many internet companies emerged during the internet era? There might be millions, but today only a few hundred are truly successful, and there are only a few dozen or even fewer that are very large. The AI industry will be the same in the future. Most AI companies will ultimately fail, but those that are truly successful may achieve extremely great success.

Kevin: But there is one difference here. Tokens are public, and anyone can buy them, so everyone can also lose money. In the early days of the internet, not everyone could invest in those startups.

Zhao Changpeng (CZ): This has both advantages and disadvantages. Conversely, would you wish you could invest when Anthropic was still very small? In any new industry, or even in any industry, most companies will fail, but the industry itself often continues to exist, and eventually, a few companies will develop very well. As for whether ordinary investors should be allowed to invest early, that's another question and can be discussed separately.

Personally, I do not agree with the so-called "qualified investor protection" logic, which is to protect you by preventing you from investing in early projects. I actually think this approach may make many people poorer, though it may not apply to everyone. If you look at many books related to the economy, one important reason many ordinary people in the U.S. can accumulate wealth is that they can participate in the stock market, and the U.S. has a very good stock market. If this opportunity is taken away, many people's wealth may be significantly reduced. Today, in many other countries around the world, local stock markets are not as strong as those in the U.S., and most people in those places cannot directly participate in the U.S. stock market. So the question is: should we allow more people to have investment opportunities, or should we restrict their participation under the guise of "protection"?

This is a debatable issue, and I don't know what the only correct answer is. But my philosophy is usually to give more people the opportunity to participate in investments while providing them with enough education, and then let them make their own decisions.

Kevin: Yes, this is actually the goal we wanted to achieve when we first created tokens, and it is the same with asset tokenization now.

Zhao Changpeng (CZ): Absolutely correct.

Kevin: If a project has market makers, KOLs, Launchpads, and exchanges behind it, can ordinary retail investors really win?

Zhao Changpeng (CZ): I think they can. The short answer is yes, but there are always many subtle balances involved. For example, advertising; you can do advertising very well or very poorly. If you make false promises in your advertising, that will certainly harm users. But "false" and "true" are not always completely black and white; there is actually a lot of gray area. You can overhype something or use vague wording to lead others to misunderstand, and there are many different situations in between.

Of course, we encourage everyone to do things the right way. To achieve this, many things need to work together, such as encouraging founders to do the right thing; educating users on how to choose good projects, teaching them how to evaluate projects rather than just looking at hype and marketing; and a regulatory framework is also needed. For example, most countries' stock markets have information disclosure requirements, but the crypto market is not yet that mature, so these things need to be combined.

However, in the long run, in any new industry, there will always be some people who are very good at hype and marketing, and some of them may gain short-term benefits at the expense of user interests. But in the long term, these projects will not last. The platforms and projects that can truly exist in the long term must be those that can build sustainably and solidly. The market itself will self-select, and ultimately, it is the market that decides.

How the number "4" became a meme in the crypto industry

Kevin: Can you explain the meme of the number "4" to my mom?

Zhao Changpeng (CZ): Are you talking about the meme of the number "4"? (laughs) I don't know if your mom understands memes, but it can be explained like this: there is a very active group of people in the crypto market, and as long as a token has culture, jokes, or memes behind it, they will trade it. They look for anything that is a little interesting or has a bit of viral potential.

The "4" meme itself has also evolved gradually. Initially, "4" should come from a New Year's plan I posted in 2023. I said that in the new year, I would do fewer things and mainly focus on education, regulation, compliance, and products, and then the 4th point was: ignore FUD, including fake news and negative narratives.

Later, I said that if I see this kind of FUD again, I would just post the number "4," and then you all help me explain what it means. The community loved this meme, and it seemed that within 24 hours, a negative report actually came out, and everyone immediately started saying, "4."

Everyone was very excited, or maybe not excited, but it suddenly became a trending thing. Later, a community member posted a photo of themselves holding up 4 fingers in a selfie, and I thought it was cool, and everyone liked it. A few weeks later, I also posted a similar photo, and it spread even further. After two years, the meme became very popular, and now there are probably thousands of memes with "4" in their names. That's how it evolved step by step.

Kevin: You turned a number into one of the most influential memes in the crypto industry. But every founder leaves scars; was there ever a time you publicly posted "4," but in your heart, you thought: could they actually have a point?

Zhao Changpeng (CZ): Most of the time when I post "4," it's because I know there are inaccuracies in that news. I don't think I've abused it so far, but others may have a different opinion.

CZ responds to "a tweet killed FTX"

Kevin: On November 6, 2022, you tweeted something to the effect of: "As part of Binance's previous exit from its equity investment in FTX, Binance received approximately $2.1 billion worth of BUSD and FTT. Due to recent exposure of information, we have decided to liquidate the remaining FTT on our books." After this tweet, FTX quickly headed towards collapse. FTX co-founder Sam Bankman-Fried later said that your tweet killed FTX. Do you think he is right in any way?

Zhao Changpeng (CZ): Absolutely not. Why? If your competitor can kill your company with just one tweet, it means you don't have a company at all. If he was running a company that I could kill with a tweet, then that company shouldn't exist in that way. So logically, that statement doesn't hold.

I can tweet about any company, and they won't collapse because of it. There are definitely many reasons why FTX ultimately fell, and I don't know all the specific reasons, but no company will directly collapse because of a tweet from a competitor, no matter how big that competitor is.

Assume today you create a very strong AI chip company; even if Jensen Huang tweets that he used to be your investor and is now ready to sell your stock or tokens, as long as your product is good, you have real customers, and your cash management is healthy, your company will not collapse because of that. The stock or token price may fluctuate in the short term, but the company itself will not disappear. It's that simple; I think anyone with a little logic can understand that a company won't just collapse because someone else tweets.

Kevin: So at that time, you had no anticipation or perhaps underestimated the impact your tweet could have?

Zhao Changpeng (CZ): I don't think my tweet had that big of an impact. I think that tweet was just one of many things. For example, CoinDesk had published an article about three or four days prior questioning their solvency. The fact is, they were not honest with customers; they misappropriated customer funds. Customers were told their funds were still there, but in reality, that money was used to buy other things, and ultimately there wasn't enough liquidity. Frankly, I didn't know these situations at the time. Moreover, our exit from the FTX equity investment was not "the previous year"; looking back today, it was about five years ago, but we still held FTT.

At that time, I felt that whether it was myself or Binance, since we were preparing to deal with these tokens, it was best to publicly and transparently tell the market what we were going to do. So to me, that tweet was more like a public disclosure. I never expected it would have such a big impact, and I even clearly stated in the tweet that we would gradually sell over a few months to avoid impacting the market price. I never know how the market will react, and I certainly did not anticipate the kind of market volatility that followed. Then Alameda's CEO Caroline Ellison tweeted about 20 minutes after I did. Many people said her tweet might have revealed more information than mine, but who knows.

Kevin: What are some facts about the week of FTX's collapse that most people still don't truly understand today?

Zhao Changpeng (CZ): I don't know what other internal information there is. To my understanding, the core issue is that they misappropriated customer funds, taking billions of dollars of customer funds to buy other things. Later, when the market fell and customers wanted to withdraw, they didn't have enough funds; that's about it.

From BNB's price drop to prison experience: CZ talks about the most stressful phase

Kevin: What was the most difficult day of your life?

Zhao Changpeng (CZ): There were actually several very difficult phases. I mentioned in my book that after Binance launched, BNB started trading publicly, and then the price fell below the ICO price, lasting about two to two and a half weeks. That period was very difficult for me, with a lot of psychological pressure, because thousands of people bought this coin and invested in this industry, and then they started losing money as soon as they bought it. Fortunately, this state only lasted about two and a half weeks. Later, dealing with the U.S. government was also a very stressful period, lasting about a year and a half to two years, so those phases were all very tough.

Kevin: You mentioned the U.S.; later, you spent a few months in jail. What thoughts did you have when you entered prison that you no longer have after leaving?

Zhao Changpeng (CZ): Rather than saying it was a belief, it was more of a worry. I was very concerned that I would be left there forever. Would they add another charge, then another, and another, until it became endless accusations? That was my biggest concern. Fortunately, those things did not happen later.

As for my case, to my knowledge, no one in American history has been sentenced to prison just for a single violation of the Bank Secrecy Act. Even today, most people in similar situations are not even criminally prosecuted, and many bank executives are not prosecuted either. Some may reach a Deferred Prosecution Agreement (DPA). Arthur Hayes is a relatively close example, and he is also a crypto guy; he ended up being sentenced to house arrest.

Kevin: And that was during the pandemic.

Zhao Changpeng (CZ): Yes, it was during the pandemic, so to some extent, it might have been a bit better for him. So at that time, I thought that might have been the most severe punishment, but that’s how things turned out.

Kevin: So when you went to prison, although you were sentenced to 4 months, you actually didn’t know when you would really be released?

Zhao Changpeng (CZ): The sentence was indeed 4 months, so theoretically, I knew the term, but you never know if they will bring another charge. Some aspects of the U.S. government did assure me that they would not do so, but the U.S. government is too large; a promise from one agency does not mean it can represent all other agencies.

Kevin: Andrew Tate's situation seems somewhat similar now.

Zhao Changpeng (CZ): I am not familiar with his case, so I try not to comment on other people's legal cases.

Kevin: When things got tough, did anyone suddenly disappear, surprising you?

Zhao Changpeng (CZ): Are you talking about within Binance, my friends, or the entire industry? If it’s my own friends, I haven’t seen that happen. Binance hasn’t either; the earliest co-founders from 2017 are still here today, none have left, so it’s a very close-knit and cohesive team. But in the entire crypto industry, many people have indeed left. There are many projects in this industry that initially promised the whole world, even promising to give you the moon, but then people disappeared a few months later.

CZ Reflects on the Presidential Pardon Process

Kevin: The U.S. president signs a certain number of pardons each year, and you later received a pardon from President Trump. What was the entire process like, from the initial contact to the final signing?

Zhao Changpeng (CZ): I think the data you mentioned might be a bit off. From what I’ve seen, many people have been pardoned or had their sentences commuted during Biden's term; I recall about 1,700 people, so it’s far more than just 40 people a year; it might average a few hundred per year. I think that should be a very high record; as for how many people President Trump pardoned, I’m not sure. In my own case, I discussed this with my lawyer, asking them how the process should go.

During Trump’s campaign, he publicly stated he would pardon Ross Ulbricht, the founder of Silk Road. His case was much more serious, involving drugs and weapons transactions on the site, and he was sentenced to life imprisonment, but later he received a pardon. Subsequently, Arthur Hayes also received a pardon. Arthur Hayes's case is much closer to mine, also related to the Bank Secrecy Act. According to what I read online, he might have been more directly involved in some trading activities; I’m not sure if that’s the exact fact, but that’s the description I saw. Later, he and several of his co-founders received pardons. Arthur Hayes was pardoned in March 2025. Before that, I had already discussed these matters with my lawyer. Later, a friend recommended a lawyer who had handled the pardons for Arthur Hayes and Ross Ulbricht, so I contacted the same lawyer, who explained the entire process to me.

Of course, some specific conversations are protected by attorney-client confidentiality, so I don’t want to disclose those details. The general process is to prepare and submit a formal pardon application. I submitted my application in April 2025, which was after Arthur Hayes received his pardon. After submission, it was just waiting; the lawyer would update me every two weeks or every month, telling me, "Still waiting." No one knows when there will really be a result.

Kevin: Did you need to meet Trump in person?

Zhao Changpeng (CZ): I have never met Trump in person, never called him, never chatted with him, never emailed him, and had no text exchanges; I don’t even have his phone number. The closest I got to him was probably in Davos. He was on stage, and I was in the audience. We didn’t shake hands, didn’t say "hello," he wouldn’t even know I was in the audience.

Kevin: Some people believe that this pardon proves that political influence is more important than the judiciary itself; where do you think this statement is wrong?

Zhao Changpeng (CZ): First of all, my perspective on this issue is certainly very biased, so ultimately, I’ll let others judge for themselves. Many people have said that the treatment I received was not fair. In my case, to my knowledge, no one has ever been sent to prison just for a single violation of the Bank Secrecy Act. Moreover, the core of the charges against me was that Binance did not establish a sufficiently strong KYC and anti-money laundering system. I personally never handled any transactions directly; I wouldn’t do those things at all; all transactions were completed through the system, and their claim was that the entire system was not strong enough in terms of anti-money laundering.

But if you look at the situation today, based on the information I currently have, the international version of Polymarket does not do KYC, while the U.S. version does KYC, but the international product is more controlled through geofencing. For perpetual DEXes, there seems to be similar discussions now; international decentralized perpetual contract products may not require KYC in the future. So different periods, different governments, and different regulatory focuses will have different standards. My perspective on this issue is obviously self-interested, so I’ll let others judge for themselves.

Kevin: So what you mean is that you encountered problems because you were doing these things too early, and the regulatory rules were not mature at that time; but now regulators need to rethink how to regulate perpetual DEX and prediction markets.

Zhao Changpeng (CZ): Yes, at that time, Binance had issues due to insufficient KYC, and I personally encountered problems because of that, but today some platforms do not have KYC at all.

Looking back with today’s standards, what I did at that time might not even be considered to have the same issues. Of course, different periods, different governments, and different focuses are all different, and that’s not for me to judge.

If X Does Payments, CZ Would Support Elon Musk to Push Forward Directly

Kevin: You previously invested about $500 million to participate in Elon Musk's acquisition of Twitter, now known as X. If Elon called you tomorrow and said, "CZ, we are finally going to do X Money, and we are preparing to put Bitcoin at the core," what would you tell him not to do?

Zhao Changpeng (CZ): I don’t know; I would probably say, "Go for it."

We were more of a passive investor at that time; I wouldn’t tell Elon what he can or cannot do. He is a very strong entrepreneur, that’s for sure, so I would say, "Go for it; it’s good for the whole industry and for the crypto industry."

Kevin: How far do you think we are from this actually happening?

Zhao Changpeng (CZ): I don’t think it’s too far. I see there are already discussions on X about accepting stablecoins; once they start accepting stablecoins, the next step to integrate Bitcoin, Ethereum, BNB, or other assets might just be a step away technically.

Bybit Hacking Incident and Binance's "Red Button"

Kevin: Last year, we interviewed Bybit co-founder Ben Zhou, just two days after Bybit experienced a hacking incident of about $1.5 billion. After the incident, you publicly suggested that Bybit pause withdrawals, but Ben Zhou ultimately did not do so. Looking back, do you think Ben was right and you were wrong?

Zhao Changpeng (CZ): I think there are no absolute rights or wrongs in such matters; it really depends on how you judge the situation at the time. Ben clearly had more internal information about the incident than I did; he would discuss with the technical team and make his own judgment based on the information they had. The final outcome did not worsen, which is certainly good; continuing to allow withdrawals did not cause new issues. Just according to my style, I would take a more conservative approach; I would pause first and then look for the source of the problem.

Kevin: Did you do that when Binance was hacked for about $40 million in 2019?

Zhao Changpeng (CZ): Yes, that’s a good example. In May 2019, Binance was hacked for about $40 million in one go, and the first thing I did was pause all withdrawals; in fact, we first stopped the entire system. In my mind, an exchange should have a virtual "big red button." Once a security incident occurs, the first thing to do is press that button to stop everything, then investigate what happened and how extensive the impact is; you shouldn’t run the system while slowly figuring it out.

In my view, anyone responsible for related systems within the exchange should have the ability to press that red button and say, "Stop, everything should stop." Binance had such a mechanism at that time, and I believe it still does. Only after you analyze everything clearly and have a very high level of certainty should you gradually restart things. Of course, doing so will affect trading continuity. Especially if you run a futures platform, market prices are changing, and liquidations that should have occurred might not happen, which could lead to many other impacts. But in my view, these are secondary issues; compared to a potential security vulnerability continuing to cause more funds to flow out incorrectly, other issues should be prioritized behind that.

In 2019, Binance paused withdrawals for about a week, but we resumed trading a few hours later because we didn’t want the prices on Binance to completely disconnect from other exchanges. In the initial few hours and days, we carefully checked the entire system. Although we quickly felt we understood the problem, I still took a more conservative approach and paused withdrawals for a week.

Moreover, during that week, we made a lot of upgrades to the system. We even adjusted the entire system architecture from a traditional server architecture to a more serverless architecture. I won’t go into too many technical details. So looking back now, you could say Ben Zhou’s judgment was technically correct because no further issues occurred afterward. But what if a second attack had happened at that time? The outcome would have been completely different.

Kevin: If Bybit had followed your advice at that time, would the outcome have been significantly different?

Zhao Changpeng (CZ): If that was the only attack in the end, there wouldn't be much difference because the attack had already happened. The situation that would create a huge difference is if a second attack occurred afterward. Fortunately, it did not. So looking back now, you wouldn't feel there was a huge difference in the outcome, but I don't know all the risk information they had at that time. It usually takes a few days to really analyze everything clearly.

Moreover, they were using a third-party wallet service provider at that time. As I understand it, the code related to the attack may have existed for a while and was not suddenly introduced that day. The hacker likely changed part of the UX or code in advance and then waited for the right moment to actually activate the attack. So what the right decision was is actually very subjective. In hindsight, we can certainly discuss many "what ifs," but such discussions have limited value. At that time, I publicly suggested pausing withdrawals, which was also an attempt to help Ben to some extent.

If he suddenly announced a pause on withdrawals himself, he would face a lot of user pressure. But if I, as a relatively influential person in the industry, publicly said, "I think pausing withdrawals might be a safer choice," then if he really decided to do so, he might gain more legitimacy from a social perspective. Users might also complain less because they would say, "CZ supports this." So I was indeed trying to help him. I proposed a more conservative plan that would bring some inconvenience to users but was safer. It was completely fine that he did not adopt it; in such cases, there is no absolute right or wrong, and it is always a very difficult judgment.

CZ on Hyperliquid: Competition, KYC, and HYPE Listing

Kevin: Many people in the crypto industry believe that Hyperliquid is the first true competitor Binance has encountered, and some think your previous public comments were an attempt to slow down Hyperliquid's momentum. How would you respond?

Zhao Changpeng (CZ): They are wrong. I hope to see more crypto exchanges and trading platforms emerge and innovate. We see exchanges as participants that help the entire industry grow the pie because today the entire market is still far from saturation. If calculated by the number of people, perhaps only 5% to 15% of people own some form of crypto asset; but if calculated by net worth, most people's wealth has only a very small portion in crypto assets. So from a funding scale perspective, I believe the penetration rate may be less than 1%, or even just a fraction of that.

Moreover, today's market size is not the future market size; the entire FinTech market should be much larger in the future. So at this very early stage, we should hope for as many participants in the market as possible to grow the industry together. Of course, there is indeed a narrative in the industry where some in the Hyperliquid community say centralized exchanges are evil, especially saying Binance is evil, and use this to attract some people who dislike Binance to build their own community. That's fine; that's their way of competing, competition is competition.

But we do not oppose them. When Trump mentioned Hyperliquid, I even publicly tweeted that it was great because it is good for the entire industry. As you can see, when HYPE rises, other assets also rise, BNB rises, Bitcoin rises, and the overall activity in the industry increases. I hope this change can become a turning point for the market to emerge from its slump.

Many people misunderstand how I view Hyperliquid; we welcome more decentralized exchanges. Moreover, I have always held the view that a pioneer in a track is not necessarily always the long-term biggest winner. Pioneers need to do a lot of work and will make significant contributions to the industry, but the first-generation products are not necessarily the best optimized. Many times, second-generation or third-generation products may perform better.

Google was far from the first search engine, Facebook was not the first social network, and Binance is certainly not the first centralized exchange. Even if you look at Uniswap, there are now other DEXs that can compete with it in trading volume. So first-mover advantage does not mean you will always win. I am not an enemy of Hyperliquid; I actually hope they continue to pave the way. What they are doing in the U.S., if it really pushes the industry forward, is very good.

There may be some people in their community who want to attack Binance, but we are not their enemies; we are open to everyone. Moreover, the more perp DEXes there are in today's industry, the more competition there is, which is better for the entire industry. We need to look at this from a broader perspective. Even though I am a major shareholder of Binance, I am also a large holder of BNB, and we have invested a lot of resources in the entire ecosystem. The growth of the industry itself is much more important to me than the growth of any single platform.

Kevin: What are the areas where Hyperliquid is currently doing better than Binance?

Zhao Changpeng (CZ): They do not have KYC.

Kevin: Hyperliquid founder Jeff was also on our podcast before, around the time HYPE Token was launched, which was about two years ago. Why has HYPE still not been listed on Binance Spot today? It has been one of the best-performing assets over the past two years, increasing about 25 times, and has entered the top market cap rankings.

Zhao Changpeng (CZ): When HYPE appeared, I had already left my management position at Binance, so I do not know the specifics. As I understand it, for a long time, HYPE was not fully freely tradable; it mainly existed within Hyperliquid's own chain and infrastructure. Initially, some centralized exchanges that listed HYPE may have needed to keep HYPE on Hyperliquid's own infrastructure or chain. Binance's requirement may have been that exchanges must be able to truly control and hold the assets they list, mainly to reduce counterparty risk. Assuming one day that project or its infrastructure disappears, Binance would at least still truly hold those tokens. I believe Binance has always had this policy, which is that Binance must truly hold the corresponding tokens. This may no longer be an issue today, but I do not know the current specifics.

After the 10/10 Crash, Why Did Binance Become the "Villain"?

Kevin: We previously mentioned the collapse of FTX, which led the crypto industry into a long bear market in 2022 and 2023. Another event happened last year that also plunged the entire industry back into a slump, which was on October 10, 2025. Many altcoins dropped to near zero within minutes, causing one of the largest liquidation events in crypto history. It seems that every financial crisis has a "villain"; in 2008 it was the banks, after FTX it was SBF, and after 10/10, an astonishing number of people chose to see Binance as the villain. Why does the market need a villain, and why did they choose Binance?

Zhao Changpeng (CZ): Let me first answer why Binance was chosen, and then analyze the situation at that time a bit more deeply. I think it is clear that a competitor was very actively pushing a narrative that the problem was caused by Binance. But in reality, if you look at the price changes, the initial changes happened elsewhere. The general sequence of events was: President Trump announced tariffs, the stock market fell, the crypto market followed suit, and then there was a relatively minor technical issue with a certain stablecoin on Binance.

Kevin: It was Ethena's USDe.

Zhao Changpeng (CZ): Yes, it was Ethena, but it was not a stablecoin of the scale of USDT or USDC; at that time, its scale was relatively small, perhaps in the tens of billions. Binance did indeed have some related issues, but that was not the cause of the entire 10/10 crash. Then the media began to attribute everything to Binance. Cathie Wood mentioned in a podcast or interview that Binance might be one of the factors. I think she may have just read the news and did not study it in depth. Later, Chinese media and a Chinese competitor that operates a centralized exchange amplified this narrative significantly.

I later recorded another podcast with Cathie Wood, and in that podcast, if you search for it, you can see that Cathie Wood clearly stated that she does not believe Binance caused the entire market crash; she only said that Binance had a technical issue at that time, which was one component of the entire crash event. If you really look at the timeline, Haseeb from Dragonfly also publicly shared a chart. After Trump announced tariffs, prices had already started to decline, and Binance's issue occurred later. So Binance may indeed have had a localized problem, but it was not the starting point of the entire market crash. The problem is that people want to find a villain, especially after many people lost money, they want to find someone to take responsibility.

Later, Binance actually put out about $800 million to compensate some users who suffered losses on the platform. Binance itself was not obligated to do this, but still did so proactively. However, such things are easily overlooked because everyone prefers to focus on negative narratives. The users who actually stayed on Binance know what happened. You can look at the social media narrative; those things can be bought or influenced, but what you should really look at is the funds. During the time when Binance faced a lot of external criticism, Binance actually still saw net inflows, and users continued to deposit more funds into Binance, and Binance's market share did not decline. So the users who truly use Binance do not believe Binance caused that problem.

Kevin: Did you post the number "4" at that time?

Zhao Changpeng (CZ): No.

I usually post "4" when some traditional media or large news organizations publish obviously erroneous negative reports. But if it is just many different people, small media, or some KOLs in the community saying it, I usually do not do that. We have seen many KOLs send me screenshots saying someone offered $20,000 or $50,000 for them to speak negatively about Binance for a month. If you search, you should be able to find quite a few KOLs publicly saying they received such offers. Some people may have really taken the money and then posted negative content, but that level is usually not enough for me to specifically post a "4."

CZ's Next Phase: Regulation, Investment, and Education

Kevin: If Binance disappeared tomorrow, what contribution do you hope to leave to the crypto industry?

Zhao Changpeng (CZ): I don't know, I just try to do what I can. I don't have a particularly fixed way to define this matter. In fact, even if Binance is still around today, I hardly spend any time on Binance, the centralized exchange itself. I spend more time communicating with governments from different countries, helping them discuss cryptocurrency regulatory frameworks, which can benefit the entire industry. I also spend a lot of time investing with YZi Labs; about 70% to 80% of our investments are still in the Web3 space, and I communicate with many founders. Additionally, I also spend some time at Giggle Academy. So whether Binance exists or not doesn't really affect my decisions on what to do next.

Kevin: What do you want to optimize the most in your life right now?

Zhao Changpeng (CZ): Positive impact. I believe that having a positive impact on others brings happiness. If you know you are making a contribution and helping others, you will feel joy internally. To some extent, humans may be genetically shaped this way. So my way of thinking is: what can I do to maximize my positive impact?

I don't have a specific number or a must-complete goal; I just hope to do my best. If I have strong abilities, I will do a bit more; if my abilities are limited, I will do a bit less. But as long as I know I have made my best effort to contribute positively, I will be satisfied.

Kevin: So looking at today, what do you think is worth investing the most time in that can create the greatest positive impact?

Zhao Changpeng (CZ): In the short term, I think it is helping different countries establish more reasonable cryptocurrency regulatory frameworks, making these countries more friendly and proactive towards innovation while also protecting consumers. This is a more tactical and short-term endeavor. In the long term, I actually believe that Giggle Academy may have the greatest positive impact because it aims to provide free education to everyone in the world and is growing very rapidly. We have only been doing this for about a year and a half, almost two years. At the beginning of 2026, Giggle Academy had about 100,000 child users, and now it has reached approximately 1.3 million. This means that in less than 7 months, the growth has been very fast and is approaching exponential. So if we look at a time scale of 10 or 20 years, I believe this could have a very significant positive impact.

Traditional education pulls people back to the average; can AI redo education?

Kevin: You mentioned Giggle Academy and education. You also have children, and we talked earlier about how to prevent the next generation from losing motivation. I have interviewed many very successful founders who are also financially well-off, and they all have young children. I ask them, how do you plan to educate your children? Most people's answer is actually: I don't know.

Zhao Changpeng (CZ): I have some views on education, and these views may be somewhat counterintuitive and could even spark significant controversy. I believe that today's education system essentially aims to cultivate an "average person," a predictable person, somewhat like the way standardized workers were trained during the industrial era. I do not mean to belittle workers; I respect any legitimate way to earn money, but I believe that the world we live in the future will be highly competitive. For example, if you are a developer, you might be competing with 10 million developers worldwide. You better be in the top 1%, or even the top 0.1%. If you belong to that very small elite, the rewards you receive could be many times greater than those of others, and even moving from the top 0.2% to the top 0.1% could result in vastly different income or opportunities.

In other words, in the top fields, just a slight lead can create a very large difference in returns, so the future requires specialization. However, today's schools often do the opposite. If you are particularly good at math but weak in English, schools will have you spend less time on math and more time improving your English, ultimately pulling everything back to an average level.

Of course, you still need to have basic language skills, and other subjects should reach a level that allows for normal living and working, but at the same time, a person should be highly specialized in one thing they are truly good at or a few things. That is the first point.

Second, I believe that today's teacher-centered education system is very costly and does not yield good average results. Of course, there are some very excellent teachers, but the average quality of teachers is not high enough, one reason being that teachers' salaries are not high enough. Many of the smartest and most capable people choose to start businesses, work in finance, technology, or become bank executives rather than becoming teachers. According to the data I understand, the average salary for teachers in the United States is about $30,000 per year, which is already considered relatively high globally, while many other countries are lower.

So we are not actually getting the most capable people in society to participate on a large scale in educating our next generation. Moreover, the entire system is also quite inefficient, with one teacher facing 30 or even 50 children, each with different levels and learning speeds. Therefore, I believe that today's education system has problems in many aspects. I actually strongly believe in an AI-driven, app-driven education model that personalizes courses for each child and delivers them through digital platforms. If done well, I believe it would be much more effective than today's system.

Kevin: Are you envisioning a digital platform, or in the future, will there be a robot that accompanies children for two to three hours a day, truly understands the child, and helps them grow one-on-one?

Zhao Changpeng (CZ): Personalization is key. In theory, both robots and apps can accurately understand a child's current state and adjust the curriculum and teaching content based on the child's situation. Robots have some advantages, such as potentially accompanying children in physical education classes and being more suitable for some face-to-face dynamic teaching. Apps also have some drawbacks, such as a lack of social interaction, but we can try to address or reduce this issue through the social features of the app. For example, we can organize children and parents together to increase offline social interaction, and we can also organize offline physical education classes or other activities through digital apps.

I believe robots will definitely be part of the future, but today, robot technology is still somewhat clumsy and has not yet reached a truly natural state like humans. We observe that young children do like robots, but once they realize "this is a robot," their interaction with the robot is not the same as with a real person. Conversely, children are actually very accustomed to interacting with apps. Even in terms of content, we have found similar phenomena. For example, we now have AI-generated children's storybooks, and today you just need to click a button to let AI automatically generate storylines and images, creating a 20-page storybook. However, we found that children do not particularly like these completely AI-generated stories. What they really enjoy is when someone thoughtfully designs the story and then uses AI to help generate images or speed up some of the processes; this kind of content performs better.

So I think that when robot technology truly matures, for example, when robots can have very natural facial expressions and engage in high-quality, very realistic conversations with people, we will definitely try it. But today, apps are much easier to scale. Once an app is well-made, whether 100,000 people use it or 1.3 million people use it, our marginal cost is almost the same, basically close to zero, only incurring a bit of extra cost. Therefore, the scalability of pure software solutions is much stronger, and this is also something I know how to do: to build apps and deliver them to hundreds of millions of people. Robots involve hardware and many other issues that I am not yet prepared to handle. But once robots truly have good facial expressions, natural control abilities, and the capacity for high-fidelity interaction with people, we will definitely adopt them.

CZ: I cannot drive the entire cryptocurrency market with a single tweet

Kevin: You mentioned earlier that you hope we have already come out of this bear market. About 25% of traders on Kalshi believe Bitcoin will break $100,000 before January 2027. Everyone knows you have always been bullish. On August 19, when Bitcoin was around $56,000 to $64,000, you tweeted: "If you want your future self to thank you for what you did today, then in the next few days, the market rose about 20%." Does CZ have the ability to drive this approximately $3 trillion cryptocurrency market?

Zhao Changpeng (CZ): Short answer: No, I do not have that ability. I have tweeted bullish sentiments too many times. In fact, as you said, I have always been very bullish; almost all my tweets lean bullish. I might tweet 5 or even 10 times a day. That tweet just happened to come about 5 hours before Trump spoke, and I had no idea that speech was going to happen. If I had known any information that could affect the market in advance, I would not have tweeted about it. Many people say that the cryptocurrency market still largely follows a four-year cycle. Some even calculate strictly according to the four-year cycle, saying that October will be the lowest point. My thought is, why do we have to wait until October to say something positive? That tweet was actually sent while I was on the road. I was a bit bored, and I usually tweet a lot every day.

I was reading a book that had a line which roughly said: If you want your future self to thank you for what you did today, you must take action today. I have forgotten who said it, but after I saw it, I thought I could slightly modify it into a more positive statement suitable for the crypto community, so I tweeted it. As a result, about 5 hours later, Trump gave a very positive speech, mentioning things like Hyperliquid, but not just Hyperliquid; the entire press conference was very positive for the cryptocurrency market. I had no idea what would be said; no one told me in advance, and I did not know any of the content.

So that's how it is. If you are always bullish and tweet 5 bullish tweets a day, there will always be a time when one of those tweets coincidentally appears 5 hours before some major positive news.

Kevin: At least today, no one is complaining about that tweet.

Zhao Changpeng (CZ): Right.

Kevin: Yes, everyone is happy today. Thank you very much, CZ, this conversation has been great.

Zhao Changpeng (CZ): Okay, thank you for inviting me.