Binance founder and Giggle Academy founder Changpeng Zhao said he would not recommend that ordinary investors repeat his 2014 decision to sell an apartment and go all in on Bitcoin if that property makes up most of their wealth. Speaking during a community Q&A at Binance Bali Clubhouse 2026, Zhao said most people should use dollar-cost averaging instead, putting 1%, 5%, or 10% of their monthly income into major crypto assets they trust.
The session was hosted by Jessica Walker and sourced from Binance. The original program was titled Should You Go ALL IN on Bitcoin? Binance Founder CZ's Answer (and Why DCA) | Bali Clubhouse 2026 Q&A and aired on Aug. 23, 2026. TechFlow compiled and translated the discussion, while also noting that the episode involved views on BTC, BNB, and sectors including RWA and DeFi, and should not be treated as independent third-party investment advice.
From an all-in Bitcoin bet to a DCA framework for most people
Jessica Walker asked Zhao whether there is a 2026 equivalent to his 2014 move of selling an apartment to buy Bitcoin, and whether he would suggest that a 25-year-old do something similar today.
Zhao said no. 「I would not recommend anyone sell their apartment and go all in if that is most of their wealth.」 He said the choice has to be personal because crypto prices are volatile, and anyone taking that route needs to understand both the risk and the downside.
He added that his own situation at the time was unusual. He had very strong conviction that Bitcoin would rise over the long term, and his daily life did not depend on Bitcoin's price. Even if Bitcoin had gone to zero, he said, he would still have been fine. Most people are not in that position.
Zhao said he still believes crypto will keep growing over time and mentioned Bitcoin, BNB, and other strong projects. But for most people, his advice was much simpler: put 1%, 5%, or 10% of monthly income into major crypto assets on a recurring basis.
TechFlow highlighted that percentage range as the most actionable part of the conversation. Using its example, someone earning $10,000 a month could allocate $100 to $1,000 into large-cap crypto assets on a regular schedule, without using leverage and without taking bets on smaller tokens. In that framing, Zhao's own 2014 trade was presented as an extreme case rather than a template.
Zhao says his thinking changed on NFTs, meme coins, and RWA
When asked what major crypto belief had changed over the past decade, Zhao listed several.
The first was NFTs. He said he did not think they would become popular, but they did. The second was meme coins. He said he also did not expect them to become so big and still does not fully understand the logic behind them.
The biggest shift, though, was in real-world assets. Zhao said that about a year and a half ago he did not see much in RWA, but now he finds the sector attractive because of features such as 24/7 trading, transparency, low fees, and global participation.
His broader point was that he does not cling to an old view once the market proves something has traction. 「A lot of things I didn't see at first, but when they became popular, I went back and tried to understand them,」 he said.
The four-year cycle still looks intact, but he cannot call the next catalyst
On market cycles, Zhao said the four-year pattern still looks strong. He noted that four years ago the market was also roughly in a bear phase, similar to now, and said the pattern has not been broken so far.
That did not lead him to a fresh prediction on what comes next. Instead, he said the next breakout catalyst is extremely hard to identify in advance. He pointed to DeFi Summer in 2020 as an example, saying he could not have predicted it six months before it arrived.
The source material named RWA, perpetual DEXs, and AI agents as areas that are currently rising. Zhao said he is not sure which of them, if any, can push the broader crypto market to new highs. 「I don't have a crystal ball,」 he said.
TechFlow's summary of that section was that Zhao did not offer a new directional market call. The more durable takeaways were two disciplines: use DCA instead of all-in positioning on the investment side, and spend 30 to 60 minutes a day on deliberate learning, drilling three to five layers deeper into each question.
On AI and crypto, Zhao favors open access and crypto-native payments
Walker also asked whether AI could become a nightmare for retail traders by trading faster than humans and spotting vulnerabilities earlier, and whether Binance should build guardrails.
Zhao said the question assumes AI will only be available to a small group of people, and he does not think that will be the case. In his view, nearly everyone will use AI. He acknowledged that AI can find vulnerabilities more quickly, but said developers using the same tools can also patch them faster and build safer systems.
He also said some users will inevitably adopt AI for trading earlier and more effectively than others. Even so, he expects the tools to spread over time, much like the internet and technical analysis software did.
For that reason, he said Binance should not try to act as a shield protecting humans from being outperformed by AI bots. His preference is to make AI tools broadly available. At the API level, he said, an exchange only sees orders arriving and cannot tell whether the decision engine behind them is human or machine. In his view, expanding access makes more sense than trying to block the technology.
On the role of blockchains in an AI economy, Zhao said the infrastructure will matter a great deal. He said AI will eventually make payments, execute trades, and book hotels on users' behalf, even if that day has not arrived yet. For those use cases, he argued, crypto-native payment systems are a better fit because they are programmable, neutral, borderless, fast, and low cost.
Credit cards, by contrast, were built for humans. Zhao said an AI using cards would run into face verification, SMS verification, and transactions that get declined at random, which makes the system a poor fit for autonomous agents.
How Zhao uses AI, and where he says it falls short
Zhao said his most common AI use case is information gathering. In the past, if he did not understand something, he would search the web, filter out ads, and read several articles before forming a view. Now he asks AI directly, which he said saves time.
He also described a failed attempt to use AI for coding. Zhao said he tried building a Gmail spam filter with AI-generated code, spent a long time on it, and still did not get a good result, so he dropped the idea.
Because he no longer writes much code himself, he said AI coding tools are not especially valuable to him personally. For developers, however, he said AI could improve productivity by 5x to 10x. Even then, he stressed that developers still need to understand every line of code AI produces and should not treat it as a black box.
On news consumption, Zhao said he does not use AI either. He prefers to look directly at what people he follows on X are posting. He said traditional news contains too much subjective opinion, so he skips it.
Giggle Academy found that fully AI-led teaching did not work for children
Walker asked how Giggle Academy combines AI, decentralized identity, and gamified education.
Zhao said more than 1 million children are already learning on the platform, which is mainly focused on content for ages 2 to 6.
He said the team tested fully AI-led teaching and found that it did not work. Chat-based AI does not actively ask children questions, he said, and many children stop after asking three or four of their own. He also said fully AI-generated content is less appealing to children, and even very young kids can sense that it was not made by a person.
As a result, Giggle Academy uses a human-led model with AI-generated components. Zhao used storybooks as an example. One-click AI-generated books were not popular, while stories written by humans and supplemented with AI-generated components performed better. He added that the team working on this effort has 60 people.
Zhao was also cautious on decentralized identity. He said it is not as important as many people think because blockchains are already very transparent. Combined with AI and KYC data from a few centralized exchanges, on-chain transactions become easy to trace. In his view, banks and cash are actually more private.
Indonesia has users, but still lacks local assets and clearer rules
When the discussion turned to Indonesia's Web3 development, Zhao said the local community is highly active but still missing several building blocks.
He said Indonesia does not have its own stablecoin. It also lacks tokenized local real-world assets, including Indonesian stocks, real estate, gold, oil, rare earth minerals, and government bonds.
To make the point, Zhao said USDT is mainly backed by U.S. Treasuries, which he described as an indirect version of tokenized government debt. He then asked what country would not want hundreds of millions of people around the world to be able to buy its government bonds directly.
He added that a stablecoin tied to the domestic currency is also important because it creates a price discovery mechanism. In his view, the gaps he listed also show that Indonesia needs clearer regulation.
His advice to a 25-year-old: ignore most specific calls and spend the time learning
Near the end of the session, Walker asked what advice Zhao would ignore if he were 25 years old today, had no money, and had no network.
Zhao said he would ignore most highly specific advice. Calls to buy one asset, sell another, build one project, or avoid another are often too tailored to the person giving them and may not fit someone else's circumstances.
What he would do instead is spend 30 to 60 minutes every day learning, with younger people putting in even more time. He said that can include YouTube, X, or asking questions through AI, but the process has to be deliberate rather than passive scrolling.
He stressed that the questions themselves matter. Rather than asking broad prompts such as why the market went up or down, he said people should keep drilling three to five levels deeper into each issue. He also strongly recommended reading, especially nonfiction and self-improvement books. One book on Amazon costs $9.99, he said, making it one of the cheapest ways to acquire knowledge.
The line TechFlow used at the top of its article came from this closing section: 「I don't know what the next big catalyst is, but if you spend an hour a day learning, the way you think will slowly change.」 Zhao said that shift will not happen overnight, but it will show up over time in deeper analysis and better decisions.
