Authors: Victor ( @vcmktasa ) · Mr. Z ( @168MrZ )

Guest: Chris Lee ( @ViewsOfChris )

In 2026, the main theme of the capital market has shifted from cryptocurrencies of the past decade to AI computing power, semiconductors, storage, and optical interconnects. 168X invited Chris Lee (李书沸), an early OG in the cryptocurrency world, co-founder of Merkle 3s Capital, and former CEO of OKEx (now OKX), CFO of OK Group, and CFO of Huobi , to share his perspective on the shift from exchanges and cryptocurrency funds to fully embracing the AI ​​theme in the US stock market.

Chris connected with Mr. Z and Victor from Palo Alto, Silicon Valley, the world's AI hub, for over an hour-long conversation. He frankly stated: "The biggest alpha right now isn't in altcoins, it's in Nvidia's financial reports, Musk's space narrative, and the supply-demand gap in AI storage and semiconductors." From position discipline, the valuation differences between optical modules and storage, the revaluation logic of HBM Super Cycle, to the possibility of Bitcoin dropping to $70,000 in the short term, the real opportunities in Web4 (Web3 + AI), and geographical and industry advice for young Chinese people, Chris gave a comprehensive review of his experiences over the past few years, which involved real money.

Table of Contents

• I. From Crypto OG to AI US Stocks: A Shift in Perspective Based on Real Money

• II. The "Wringing Out a Towel" Correction in US Stocks: A Healthy Correction in a Bull Market vs. The Real Signal of a Bubble Breakdown

• III. Position Discipline During the Main Uptrend of AI: Why 20-30% Cash is "Offensive" Rather Than Defensive

• IV. Optical Modules vs. Storage: Valuation Differences, Smart Money, and Business Timing

• V. HBM Super Cycle: From Cyclical Stock to AI Pricing Weighting in Storage

• VI. Can Chinese storage technology challenge the three giants? What are the limits of CXMT and YMTC?

• 7. Why invest in US stocks instead of cryptocurrencies: The biggest alpha isn't in altcoins.

• 8. Bitcoin Technical Analysis: Potential dip to the 65,000-70,000 range

• 9. Two conditions for the next bull market in crypto: AI phase peak + new institutional entry.

• 10. Crypto's Historical Mission: From Casinos to Financial Infrastructure

• XI. Web4 = Web3 + AI: Stablecoins, Cross-border Payments, and AI Agents

• 12. Nasdaq Correction Path and Conditions for the Return of the US Stock Market Bull Market

• Thirteen, Geographical, Industrial, and Skills Choices of Young Chinese Americans in the AI ​​Era

• Fourteenth, final advice to the younger generation: Choice is more important than effort; peace of mind, restful sleep, and a safe home.

I. From Crypto OG to AI US Stocks: A Shift in Perspective Based on Real Money

Mr. Z: Could Chris please introduce himself briefly? How did your investment perspective evolve from focusing on exchanges and the cryptocurrency market in the early days to now paying close attention to US stocks, AI semiconductors, and storage?

Chris: Hi everyone, I'm currently in Silicon Valley, in Palo Alto, the heart of the AI ​​universe. It's evening here now.

I joined Crypto in 2016. I started at OKX, which was then called OKCoin and OKEx; later I went to Huobi as CFO; then I went to Black Hole Capital as president. Black Hole Capital was one of Binance's earliest investors. CZ and He Yi were also originally from the OK group. I met them in 2017 and have maintained a very good relationship and contact ever since, and we have met individually over the years.

I started my own fund in 2020. After leaving Blackhole, I founded my first fund called TKX Capital, which invested in about 70 or 80 crypto projects, and of course, I learned a lot in the process. Now, I have my second fund, Merkle 3s Capital, which mainly focuses on "Web4," that is, Web3 + AI . We also have a secondary AI-themed fund, managed by another partner, which has performed very well this year.

The shift in perspective isn't as dramatic as people might imagine. I've always believed that the only factor determining price is the flow of funds. For the past decade, funds have been flowing into cryptocurrencies, which is why we focused our exchanges and funds there. But in the last two years, it's clear that funds are flowing into AI computing power, storage, and optical interconnects.

Coming from a traditional professional management background, my approach is essentially to follow the flow of real money. This is crucial. But please don't misunderstand, I'm not abandoning cryptocurrencies, but rather acknowledging a reality: the biggest alpha right now isn't in altcoins, but in Nvidia's financial reports, Musk's space narrative, and the supply-demand gap in AI storage and semiconductors. I'm still involved in cryptocurrencies, just with an added AI perspective, nothing more. In short, it's about following the flow of funds.

II. The "Wringing Out a Towel" Correction in US Stocks: A Healthy Correction in a Bull Market vs. The Real Signal of a Bubble Breakdown

Mr. Z: You recently mentioned that after six consecutive weeks of gains, US stocks have entered a "squeezing-out" correction. How do you interpret the current market situation? Is this a healthy adjustment within a bull market, or the beginning of an AI bubble bursting?

Chris: Let's start with the AI ​​narrative itself. Ren Zhengfei of Huawei once said that the AI ​​revolution is the last industrial revolution in human history because it will change everything. So the main theme is actually very clear: AI is the "electricity revolution" of this era . Think back to railways, highways, and automobiles; they were all things of the same scale.

My assessment is that this is a healthy correction within a bull market, not a bursting bubble. When I said, "After six consecutive weeks of gains, it's time for a healthy pullback," the market did indeed pull back for several days, but has now rebounded. In any case, I think there should be another healthy pullback before further upward movement becomes more stable.

The characteristics of a bubble bursting are: orders disappearing, CapEx order cancellations, and downward revisions in guidance. But what we are seeing now is exactly the opposite: Nvidia, Broadcom, and TSMC's backlogs are already booked into 2026 or even 2027, and Hyperscaler's CapEx is still increasing. The side effect of six consecutive weeks of short-term gains is simply that valuations have outpaced fundamentals. The market needs to shake out overly excited retail investors and highly leveraged funds; this is called "squeezing the towel."

The real risk signals are actually these three things: First, AI earnings reports start to miss guidance; second, channel clients' capEx is lowered; and third, interest rates spiral out of control again. Before these three things occur, I would consider every major market drop an opportunity to buy.

My initial expectation was that QQQ would pull back by less than 10%, but the market is too hot right now, and that might not be the case. Add to that the Iran war, inflation, US debt, and other factors, and there might be another correction. Personally, I'd actually prefer to see such a correction; a bull market without corrections is actually more dangerous.

III. Position Discipline During the Main Uptrend of AI: Why 20-30% Cash is "Offensive" Rather Than Defensive

Mr. Z: You've always emphasized that "discipline comes first in the investment world," and you've also reminded people not to be fully invested, not to use high leverage, and to keep 20-30% or even more in cash. How do you view position management now? In the main upward trend of AI, why is cash actually an offensive tool?

Chris: To be honest, I've learned a lot from my own experience with the US stock market. My returns were excellent in the first half of last year, but in the second half, especially after October, I basically gave it all back. This experience made me realize that investment discipline and leverage control are the most important factors.

I've mentioned Munger's quote before: "Leverage, women, and alcohol" are the three things that can bankrupt you, with leverage being number one. Look at the cycles of 2018 and 2022; too many smart people were forced to liquidate their positions. It wasn't that they were wrong; it's just that they couldn't survive until dawn.

So my usual strategy is to allocate about 60-70% of my portfolio to aggressive positions, 20-30% to cash, and the remainder for hedging. Why is cash important? Mainly because having cash on hand gives me peace of mind and makes me more resilient; also, if the market corrects downwards, I have money to buy at the bottom.

In the main upward phase of AI, why is cash an offensive asset? Because the most valuable thing in a bull market isn't stocks, but the cash you hold on the day of the big drop. You see, when it hit its lowest point on March 31st, I was adding to my positions on Twitter, adding and adding until the rebound, and that wave of returns was very good.

If you're fully invested, you won't have a chance to buy at the bottom when the market corrects. You might also be using leverage, and every interest payment will leave you feeling uncertain. If the stock price falls further, the pressure will be immense. Investing is also a psychology skill; don't put yourself under this kind of unspoken pressure and make irrational decisions. If you really must use leverage, I think it should be no more than 10% to 30%. Any more and you'll forget about discipline.

As for what I'll buy during a market crash, it will be based on the main theme of AI. Nvidia represents computing power; TSMC is one of my largest holdings because regardless of whether CPUs or GPUs prevail, it ultimately depends on TSMC, the manufacturer. In fact, I think TSMC is the crown jewel of the entire AI revolution, like the Star of Africa in the Queen's crown. Then there's memory; my most important holdings previously were Nvidia, TSMC, Micron, and the DRAM memory ETF.

Incidentally, our secondary AI fund has achieved a year-to-date return of approximately 40%, but currently maintains about 50% of its portfolio in cash. This demonstrates the shared philosophy and alignment among our partners.

IV. Optical Modules vs. Storage: Valuation Differences, Smart Money, and Business Timing

Mr. Z: You've recently repeatedly mentioned that "optical modules are overpriced, while storage is relatively cheap," even suggesting that smart money may already be making its choice. How do you determine whether an AI field is overvalued or undervalued? What is the biggest valuation difference between optical modules and storage?

Chris: I judge valuations based on three things: First, is the fundamental data genuine? Second, has the consensus already been priced in? Third, are there any unmet expectations? This last point is very important.

The optical module market has already seen a significant surge, with forward P/E ratios projected to 2027 and 2028. Market expectations are very high, making this a clear signal to sell. Another observation is the 13F filing of Situation Awareness LP (an AI fund founded by former OpenAI employee Leopold, with assets under management exceeding tens of billions of dollars) . They sold off their entire stake in optical module stocks like Lumentum (LITE) and Coherent (COHR), but bought Corning (GLW).

In addition, the business timing of CPO (Co-packaged Optics) products is quite uncertain, and the market will offer a discount for uncertain things.

But storage is completely different. Look at HBM's price increase and overflowing orders, but the market habitually values ​​Micron, SK Hynix, and Samsung as cyclical stocks, and recently gave them forward P/E ratios of 5-7. If we really consider a super cycle, it should be given at least a P/E ratio of 15-20, so this is just the first wave of valuation jump.

Let's look at another statistic: SK Hynix's net profit surpassed Samsung's for the first time in 2025, thanks to HBM. Both are benefiting from AI, but optical modules rely on narrative premiums, while storage relies on pricing weighting. Moreover, seeing is believing; the numbers are there, and the expectation gap for future market performance will be even greater.

Smart money has been moving all along. Micron's price has undergone two rounds of consolidation: the first surged to 465, then fell to around 313; later, it consolidated again, rising to over 800; then it dropped to the 706 level but couldn't hold, falling further before rebounding to the 600s; and now, with 818 acting as resistance, it broke through that level immediately. Large investors haven't been selling off at all, because everyone knows this.

Having worked in the real economy myself and served as a CFO of a listed company, I know that a true revaluation of a stock involves at least several factors: the company's fundamentals (firm-specific), the industry itself (sector), the overall market (beta), and how the market views the stock (what biases they apply). The most crucial factor is when net profit growth stagnates but the valuation multiple changes. You see, Morgan Stanley and JPMorgan Chase have already started this revaluation process, and UBS followed suit yesterday; there will be more to come.

Moreover, production capacity will not be able to meet industrial demand until 2028. I've read a lot of research, and even some Taiwanese storage distributors and key figures in the value chain believe that it may not be able to meet demand even by 2030. There are still people on Twitter saying that storage is a cyclical stock, a commodity. Everything is a commodity, but look at the data growth, the valuation is still suppressed by cyclical stocks, I think that's too stubborn.

Keynes famously said, "In the long run, we are all dead." While we are all ultimately dead, the high reliability of data within the last three years provides very good information. Just accept the numbers; there's no fooling the numbers.

Furthermore, there's the factor of scale; the scale of storage modules built up over two or three years far exceeds that of optical modules. When buying stocks, you need to consider two timings: Business Timing, which is when the industry starts to scale up; and Market Timing, which is when to enter the market. These are the two points that the best investors on Wall Street consider.

V. HBM Super Cycle: From Cyclical Stock to AI Pricing Weighting in Storage

Mr. Z: You mentioned that Anthropic's explosive growth further validates the narrative of a structural shortage in AI storage. Considering the growth of Claude, Vibe Coding, and enterprise-grade AI tools, how will this translate into pricing power for HBM, DRAM, NAND, and the three major storage companies?

Chris: First of all, computing power is constantly growing, and it simply can't be stopped now. Google CEO Sundar Pichai once said that the demand for computing power roughly doubles every six months. So I don't see the moment when it will stop.

Moreover, everyone's life is already inseparable from AI. When you wake up in the morning and use Google search, Gemini's AI function provides the results directly; in mainland China, there are apps like Doubao and Qianwen, which have become part of daily life. The demand will gradually grow to a very large scale, requiring massive computing power.

The five-layer architecture I'm looking at is: energy, computing power (chips), storage, applications, and cloud (cloud services). Of course, the GPU layer is the most crucial. My current assessment remains that hardware comes first, including Nvidia, TSMC, Intel, AMD, and storage – all fall under this hardware-centric approach.

The software sector is also slowly picking up, and I see great potential in the cloud: Microsoft has now established a firm foothold; Amazon is also quite strong; and Google, if you held it from the low of 142 all the way to the current 380-something, is quite stable. Second-tier stocks like NBIS (Nebius) also have investment value when the overall market is trending upwards.

But the main focus remains on hardware, which includes CPOs and optical modules. Currently, the most attractive, most tangible, and still under-reconstruction area is storage. This is my core assessment.

VI. Can Chinese storage technology challenge the three giants? What are the limits of CXMT and YMTC?

Mr. Z: What are your thoughts on the medium- to long-term catch-up efforts of Chinese manufacturers like CXMT (Changxin Memory) and YMTC (Yangtze Memory) against the three major memory giants? Why do you believe that the supply of high-end HBM will still be highly dependent on Micron, SK Hynix, and Samsung in 2026–2027?

Chris: Chinese manufacturers must face reality, but we shouldn't idolize them. Changxin is about three years behind the three giants, while Yangtze Memory has about 11% of the global NAND market share and is also pursuing an IPO, which is quite good. But doing well doesn't mean they can do HBM well; these are two different things.

The technological competition in HBM mainly focuses on three aspects:

First, TSV (Through-Silicon Via) technology; second, advanced packaging; third, co-design with GPU manufacturers.

The first two require EUV and advanced packaging equipment, but these are currently subject to export restrictions to mainland China. The third point requires deep integration and co-design with Nvidia and AMD, which Chinese manufacturers cannot currently access. Therefore, in 2026–2027, the supply of high-end HBM3 and HBM4 will definitely remain monopolized by these three giants.

These two Chinese manufacturers will likely erode the low-end DDR and consumer-grade NAND markets in the long run, but they won't be able to make a move in the high-end AI market in the short term. This is the core logic behind my heavy investment in Hynix, Samsung, and Micron.

7. Why invest in US stocks instead of cryptocurrencies: The biggest alpha isn't in altcoins.

Mr. Z: You were a very important figure in the early days of the crypto world, but recently you seem to be focusing more on the AI ​​theme in US stocks. What are your thoughts on the current phenomenon of "investing in US stocks but not crypto"?

Chris: I think the myth of 100x or 1000x coins is really rare these days. Getting rich quick with meme coins only happens when there's high liquidity, a surge in altcoins, and a wave of exploitative practices. It's not as glamorous or outrageous as it used to be.

If you want to get 100x coins, you can only hold them if you bet on the right sector and bet correctly, and most altcoins will eventually go to zero.

We actually had a formula for successful altcoins back then: community consensus + technical consensus + team + capital (VC resources) + exchange support. Here I want to emphasize that centralized exchanges remain extremely important to this day, and that's undeniable, which is a shame. But even culture and community consensus will gradually erode. Look at ETC and EOS from the past—who talks about them now?

If you still want to get Alpha in the crypto world, it's really difficult. Forget about trying to get into low-quality cryptocurrencies; that era is over. Besides, in the past, it was all about doing the work for institutions, who picked all the lowest-hanging fruit.

But I think there are still a few areas where there is potential:

First, infrastructure and services. Institutionalization is inevitable; exchanges can't handle this. Custody compliance, stablecoin issuance, on-chain data, and AI tools—if you have connections in these areas, you still have a chance. Retail investors will find it very difficult to compete.

Second, treat cryptocurrencies as financial infrastructure, focusing on real-world business and long-term goals, not as a gambling den. Three to five years of dedicated work will yield far greater returns than going all-in on altcoins.

Third, buy Bitcoin regularly in fixed amounts and add to your position when prices drop, using Beta to beat Beta – this is the simplest thing to do.

We are still investing in some AI projects, but we need to have Web3 elements and approach them from a Web3 perspective. We are still trying in this area, and often it is also because of passion.

8. Bitcoin Technical Analysis: Potential dip to the 65,000-70,000 range

Victor: Regarding Bitcoin, you previously mentioned that it's highly likely to drop below $70,000, and if it does, it's a good time to buy. With the influence of the AI ​​era, do you think the positioning of BTC as an asset has changed? As fund managers, how do you allocate it in your asset allocation? What are your views on the future market trend?

Chris: Let me start with the technical aspects. Bitcoin is currently around 75,000, having fallen below the 200-day moving average (around 80,000), which is a key technical signal indicating a weakening medium-term trend. My personal assessment is that a drop to the 65,000-70,000 range is possible, to shake out leverage and market sentiment. Therefore, my current assessment of Bitcoin is primarily based on technical analysis.

Next, let's talk about positioning. Here, I want to correct a common assumption: many people believe that Bitcoin will only rise if funds flow back from semiconductors, storage, and AI to Crypto. This implies that the total amount of funds hasn't increased. But that's not actually the case. The total market capitalization does grow, and BTC and tech stocks are fundamentally different entities; they shouldn't be equated directly.

So what exactly are the roles of BTC and ETH? Bitcoin's role remains that of a store of value, a decentralized digital gold used to hedge against inflation; Ethereum, on the other hand, is a decentralized world computer, the underlying infrastructure of blockchain, more like digital oil. The two assets are fundamentally different. Therefore, when incorporating them into asset allocation, they should be viewed in a tiered manner, like fund management, rather than treated as the same thing.

9. Two conditions for the next bull market in crypto: AI phase peak + new institutional entry.

Mr. Z: For Crypto's liquidity to truly improve and for the next round of market activity to begin, what conditions do you think need to be met?

Chris: Two conditions need to be met:

First, AI itself has reached a temporary peak, so it's necessary to take profits and exit the market before looking for a sector with higher beta.

Secondly, new institutions are entering the Crypto space, and new trends are emerging: for example, the launch of genuine stablecoins, significant new growth in USDC and USDT (previously, before predicting Bitcoin price movements, the first thing to look at was the new growth in USDT), and RWA's growth depends on a genuine surge in the US stock market, followed by Bitcoin reaching new highs, creating FOMO. This is a chicken-and-egg logic.

Before these events occur, the crypto market is essentially consolidating, and I strongly advise against trading altcoins. Among mainstream coins, I'm most optimistic about BTC, ETH, Solana (SOL), BNB, and Hype; you can consider these options.

For altcoins to rise, the only support is increased liquidity, because they are essentially "long-tail assets." Therefore, investment should be approached with a fund management mindset: first allocate to stocks, then to cryptocurrencies, and within cryptocurrencies, further divide into Bitcoin, Ethereum, and mainstream coins. Treat altcoins as a small, recreational investment, using only a very small percentage.

Having worked in the cryptocurrency exchange industry for over five years, including the world's two largest exchanges, I've witnessed far too many instances of investors being exploited. Altcoins rarely yield long-term returns, and often you simply don't have enough time to realize them.

On January 20, 2020, I had dinner with Warren Buffett. The moment Bitcoin was mentioned, Buffett became inexplicably furious. He's 90 years old, but there's one thing about him that I still think is reliable: he doesn't invest in things that can't be accounted for or that don't have cash flow.

However, I believe Bitcoin still holds its place, primarily supported by global consensus: everyone believes in Satoshi Nakamoto and blockchain technology. Altcoins lack this consensus, so if you were to invest in altcoins, it would be with a very small percentage of your portfolio.

10. Crypto's Historical Mission: From Casinos to Financial Infrastructure

Victor: Crypto's wildest days are over. What advice would you give to newcomers to the crypto industry on how to earn Alpha within the industry?

Chris: To be honest, the Web3 era is over, and we have to accept that reality. I really admire Mr. Z and Victor, you two from Taiwan. I see your perseverance, your boldness, and your ability to go international. But even so, I think the opportunities in the crypto world now mainly belong to institutions; there are really not many opportunities for individuals.

Think about it: the most profitable exchanges, those that might earn $2 billion in net profit annually, are all laying off staff and preparing for a downturn this year. Many large exchanges may even lose money this year. Given this environment and so many restrictions, starting a startup in this industry is actually very difficult, because fundraising is also challenging.

We've invested very little in VC this year. Last year, we tried our best to help several projects raise funds; I even opened a personal check and used my own money to support them because we still believe in the rise of Chinese teams and that Chinese people are better at fundraising and storytelling. Hong Kong, such a small place, used to have big companies like Crypto.com , FTX, and Bitfinex, as well as the excellent TradFi, but unfortunately, opportunities are becoming increasingly scarce.

Furthermore, I must frankly say that many regulations in Hong Kong over the years have been unfriendly to the entire industry; legislation was implemented without proper consideration. There's a Cantonese saying, "Cut off your toes to avoid sand worms," ​​which means cutting off your toes to avoid sand worms. None of Hong Kong's 12 licensed exchanges are profitable, and Binance still has the largest user base. The same goes for stablecoin regulations; many things that should have been done haven't been done, which involves a lot of political factors.

Therefore, I wouldn't even recommend anyone to invest in Crypto right now. I think Crypto has reached the historical stage it should be fulfilling. What remains are payments (Stablecoin), RWA, and some AI-linked models. That's why I've always advocated the "Web4 = Web3 + AI" direction.

If you want good returns, I suggest primarily investing in AI, but still holding onto Bitcoin. That's my advice. Many veteran crypto investors on Twitter are still using outdated thinking to approach the current market, and most of them are losing a lot of money. We must accept the changing times.

Don't get me wrong, I'm still very bullish on Bitcoin, very bullish on Ethereum, and also bullish on one or two exchange tokens. But the biggest growth market is already in AI.

Let me tell you a little story from my own experience. A while ago, during the week of the Hong Kong Web3 Festival, I sat face-to-face with Vitalik Buterin (Ethereum founder). I asked him directly, "If you could do it all over again now, would you choose AI or Web3?" He honestly answered, "I might actually choose AI." But he added, "Staying in Web3 could also make a difference, allowing me to do something more meaningful." So even though AI offers more opportunities, if you can find a real pain point in Web3 and solve it, that's also an opportunity. Entrepreneurship requires perseverance; it's not something that can be achieved overnight.

XI. Web4 = Web3 + AI: Stablecoins, Cross-border Payments, and AI Agents

Victor: So, in your opinion, what specific areas within the Web4, or Web3 + AI, would be of interest to you and where you would like to invest?

Chris: It's actually very clear now. US regulation is very clear: the Clarity Act has been passed, and stablecoin legislation has been completed, but dollar liquidity has not really eased, and there is actually a lack of applications and new narratives.

I'm looking forward to a few things:

First, stablecoins are truly penetrating cross-border payments, becoming a vehicle for the export of the US dollar. This could even extend to stablecoins for the New Taiwan Dollar and the Renminbi. Payments are crucial; if you pay attention, you'll see that all large companies eventually enter the payment sector. Payments offer incredibly strong user stickiness, permeating almost every aspect of your life, without necessarily requiring a monopoly.

Secondly, the payment, settlement, and credit between AI agents are something that only Crypto can do, a scenario that Web2 cannot replace, and it is also more reasonable from a blockchain perspective.

Third, Nasdaq and the NYSE both want to implement on-chain RWA. Therefore, we are also looking at this direction, using tools and accessibility tools to connect the scenarios.

12. Nasdaq Correction Path and Conditions for the Return of the US Stock Market Bull Market

Victor: You recently predicted that the Nasdaq might pull back by 10-13%, and the semiconductor sector by an average of around 20%, before rebounding during the next earnings season. What is the core basis for your judgment on this path?

Chris: Actually, the Nasdaq rebounded quickly after its first pullback. I expected it to pull back another 10%, but it didn't. It only pulled back a few points and then hit a new high. So this pullback hasn't officially started yet.

The worst-case scenario is this: CTA strategies now account for about 60-70% of normal trading volume in the US stock market. They are bullish on factors that trigger price movements, such as Micron's price dropping to the 6s after breaking the support level of 706, but then breaking through the resistance level of 818. It's no longer about fundamentals; it's a momentum trade.

I think a pullback of less than 10% is more reasonable, and some small-cap stocks outside the main theme may fall even more. For example, optical module stocks may only have bottom-line growth of 30% to 40%, but their P/E multiples are already at 50-60 times, and many companies' CPO products have not even been officially launched yet. This valuation is unreasonable.

Therefore, I have always advised everyone: first, do not use leverage; second, keep some cash on hand, 10-20% is probably reasonable.

Mr. Z: For the US stock market to truly return to bull market conditions, what conditions do you think are necessary? Is it continued better-than-expected AI earnings reports, the return of interest rate cut expectations, stable inflation and oil prices, reduced geopolitical risks, or a resurgence of funds into Mag7 and semiconductor stocks?

Chris: There are really only two core points: First, AI's financial reports continue to exceed expectations; second, interest rates are no longer out of control. Out-of-control interest rates are simply inflation.

The US national debt is currently around $37-38 trillion, increasing by $1 trillion every 100 days. This debt pressure means the Federal Reserve needs to maintain low interest rates for a long time; otherwise, the government's fiscal policy will not be able to sustain itself. Therefore, interest rate expectations will fluctuate greatly, but the overall trend is downward. Inflation and oil prices will depend on the Middle East and OPEC in the short term, but in the long term, they will depend on the productivity gains brought about by AI, which is crucial.

Since Trump's second term, people have probably gotten used to it: befriend volatility, geopolitical risks are always present. Look at his actions against Venezuela and Iran two years ago; he hasn't even touched Cuba yet. These kinds of things will always exist. The market will just digest them repeatedly.

My core bet boils down to this: AI CapEx won't fizzle out, interest rates won't skyrocket, and the US stock market bull run will continue. Every pullback is an opportunity. My basic logic is Mag7 + semiconductors + memory, which are the leading sectors in this round.

Thirteen, Geographical, Industrial, and Skills Choices of Young Chinese Americans in the AI ​​Era

Mr. Z: You previously mentioned that "money only flows to open economies," and also that for young people from Hong Kong and Taiwan who grew up with British and American education, the United States remains a better place for career development. What advice would you give young people regarding location, industry, and skill set?

Chris: I just had dinner tonight with several influential AI investors from Hong Kong in the Greater Bay Area. One of them is a CEO of a listed company who has invested in several major Hong Kong exchanges. He himself feels that many policies in Hong Kong are still not open, including Web3 policies. JD.com and Ant Group withdrew their stablecoin applications, and in the end, it became two foreign note-issuing banks that are working on it.

Some people who study for their undergraduate or master's degrees in the United States have the goal of staying there; some who were originally in Singapore have also left, feeling that Singapore is not suitable for them. The most important thing is how you define yourself.

To be honest, from the perspective of both Hong Kongers and Taiwanese, we could have acted as Super Connectors in the past, especially Hong Kongers. But now, due to political reasons, this model has been weakened in both Hong Kong and Taiwan; Taiwan, also for political reasons, has blocked a lot of capital and cooperation that should have come in. This is a reality we must face.

I suggest thinking about it from three levels:

The first layer is geography. For young people from Hong Kong, Taiwan, Singapore, and Malaysia who have received British and American education, the main geographical battleground is the San Francisco Bay Area, New York, Silicon Alley, Austin, and so on. This is supplemented by Asian hubs: Singapore, Hong Kong, and Tokyo. However, there is one thing that mainland China cannot do elsewhere: the supply chain. Supply chain and R&D functions are possible in mainland China.

The second layer is industry. AI, semiconductors, energy, biotechnology, and finance are the five sectors that will be the largest pool of funding in the next decade.

The third layer is capabilities. For hard skills, choose one, such as AI engineering, product development, investment research, or sales, and aim for the top 10%. For soft skills, you must have English proficiency, storytelling ability, cross-cultural communication skills, and capital market knowledge.

The most valuable things for young people are time and the opportunity to learn from mistakes. For people like us who have been in corporate management for 20 years, one mistake can implicate our partners and colleagues, who all have families to support. This is different.

I think young people need to step out of their comfort zones, not tie themselves down, and choose the right direction. Huobi had a motto back then, the first line of which was "Choice is more important than effort." Think it through first, and then put in the effort. Otherwise, if you choose the wrong place or the wrong industry, no matter how hard you try, it will be useless.

The advantage of the new generation is that it doesn't necessarily depend on where they were born or their background. The rise of Bitcoin and Ethereum in 2012 and 2013 was, frankly, driven by the Chinese market; later, Wall Street embraced or even controlled Bitcoin; and then came the new opportunity: AI. The hardware infrastructure for AI will likely be in place until at least 2029, and we're only halfway there now.

As for other aspects, such as software, Tom Lee has been talking about software recently, and there are indeed many things to try. Do you want me to tell you how to proceed, Chris? Honestly, I don't know, I'm not sure either. We'll all explore together, but at least the direction of AI is correct.

I'd also like to specifically mention Taiwan. Taiwan is incredibly strong in hardware. Many of my Taiwanese friends come from hardware backgrounds, and there are quite a few Taiwanese hardware analysts at Morgan Stanley, all of whom are outstanding. Taiwan also excels in POS (POS) technology, with companies like Castles Technology. Add to that AI Smart Devices, and at the recent major exhibition in Chicago, I saw that globally, there are really only two places that can produce true smart device hardware: mainland China and Taiwan. There are truly no other places. Moreover, many manufacturers in mainland China are actually backed by Taiwanese companies.

Let me give you a more direct observation: There are tons of beautiful women graduating from National Taiwan University, and what industries are they working in now? Basically, it's all AI. What are those people who were at the height of Web3's popularity in Taiwan doing now? Also, it's all AI. This is the passing of an era, the change in roles and industries. But what can we possess, and what can we do to excel in the new field of AI and Web3 integration? That's what we can do.

Fourteenth, final advice to the younger generation: Choice is more important than effort; peace of mind, restful sleep, and a safe home.

Mr. Z & Victor: Finally, Chris, is there anything you'd like to say to our listeners, or to the builders and investors in the crypto and AI communities?

Chris: To be honest, I rarely do interviews on Space anymore because I feel that most of the feedback on Twitter is unthinking and mostly just insults. So for a while, I stopped using any Spaces. But I once wanted to be friends with everyone, and I was moved by the spirit of the two of you, so I'm here to talk about it today.

I think young people shouldn't change themselves to cater to the market; they should have faith in what they do.

If you were to invest your time and capital in three things:

First, allocate hardware assets. Keep a portion of your portfolio in Bitcoin. The main themes in US stocks are Mag7 + Micron, SK Hynix, and DRAM ETFs. Palantir (PLTR) is also good, but it's too expensive now. Keep cash on hand. Huge returns come from waiting and patience. Truly good companies may require holding them for two or three years without selling to see real compound interest.

Second, improve yourself. Use large AI models more often, and use GPT more frequently.

Third, building a network of connections. Mr. Z did this exceptionally well. Perseverance is crucial. Over the years, I've helped several young people, providing them with significant resources, but most disappointed me. Some projects even sabotaged investors, causing them to lose money, so I'm quite reluctant to help them. However, seeing Mr. Z's sincerity and spirit, I see hope for a new generation of young people on the island of Taiwan.

What skills will become increasingly valuable in the future? Judgment, taste, capital allocation ability, the ability to build trust, and cross-cultural execution capabilities. For example, connecting the supply chain from mainland China to the US market—this is cross-cultural execution. These are the kinds of interpersonal relationships that AI cannot replace.

If you're learning about investing, the first person you should definitely study is Duan Yongping. If you have time, take a look at Buffett and Munger; they truly demonstrate the principle of simplicity. You must define your own circle of competence, stay true to yourself and the facts; only by being grounded can you weather economic cycles.

I don't talk much about stocks because I only research a few, but I spend almost all my time researching those few stocks. Later, when I saw the problems with Micron, I used AI to investigate Micron thoroughly every day until I had a complete understanding before I dared to invest heavily in it.

As the younger generation, the most important thing is to be healthy, exercise, and take good care of our families—this is our responsibility. "Man proposes, God disposes," so just do your best. Think about your life with the mindset of fund management, and consider your life, business, and investments holistically.

Accept that you are imperfect and have room for improvement, and reflect on yourself each time. Humbleness is very important; only by humbly acknowledging your shortcomings can you find room for growth. American education often emphasizes "I'm so good, I'm damn good," but in reality, it's the opposite of mindset that will lead to progress.

Many people constantly express themselves and talk a lot, but that's not the point. Substance over form; your content is more important than your appearance. Look at Jensen Huang of Nvidia; he was originally an introvert (I-type), but gradually trained himself to become an extrovert (E-type). Elon Musk is also an introvert (I-type); so is Mark Zuckerberg. Only those who truly get things done can stand out.

The wheels of time keep turning, and we accept this change and adapt to this era. Huobi had four guiding principles back then, which I'd like to share with you:

First, choice is more important than effort.

Second, life cannot be ruined by waiting. Don't feel that you need to plan everything perfectly before you do something; opportunities don't wait for anyone.

Third, mental strength can overcome obstacles. If resources are insufficient, rely on mental strength, resilience, and hard work to achieve your goals.

Fourth, equivalent exchange. What you can offer in return for your achievements is important.

Later, I added two sentences myself:

Fifth, let professionals do what they are good at.

Sixth, three perspectives: a global perspective, a holistic perspective, and a future-oriented perspective when thinking about problems.

Finally, I want to say that in this era of highly volatile geopolitics, we must consider "what is our place to settle down and make a living?" I'm a blogger who rarely speaks the truth on Twitter. Everything has its pros and cons; mainland China has its good and bad aspects, foreign countries have their good and bad aspects, and Taiwan is no exception. Let's look at things objectively.

There are many times when politics is beyond our control. As individuals, we truly cannot control it, so we should focus on doing our best and not let politics affect interpersonal communication. We are all Chinese, and our culture is rooted in Confucianism, propriety, wisdom, trustworthiness, benevolence, and righteousness—these are our greatest commonalities.

Investing requires achieving "peace of mind, restful sleep, and family security" : peace of mind means being stress-free and able to make rational decisions; restful sleep means being able to sleep soundly; and family security means ensuring the well-being of your family. Two weeks ago, I liquidated all my Bitcoin holdings in a fund, selling everything between 81,000 and 82,000. I couldn't do this before, but because I can sleep soundly, avoid leverage, and have cash to buy on dips, I can now. Often, if you go to extremes, adding more and more leverage, just one failure can wipe you out.

In this era of AI and amidst the vortex of geopolitics, we are all ants in the grand scheme of things. But what we can do is take responsibility for ourselves, take care of our families, lead our teams well, and then pass on our hopes to the next generation. Let the next generation be smarter, and solve cross-strait issues, productivity issues, and issues of greater human harmony.

Keep striving, keep running, young people, the future belongs to you. Thank you.

Mr. Z & Victor: Thank you so much, Chris, for your insightful sharing over an hour today. From your perspective as an early OG in the crypto world to your current assessments of AI in US stocks, the HBM Super Cycle, and Web4, as well as your life advice for young people, it was all very helpful. We look forward to meeting you in Silicon Valley or Hong Kong sometime. Thank you also to every listener who has listened this far. If you enjoyed this episode, please subscribe to 168X on WeChat and YouTube, and share the show with more friends interested in macroeconomics, AI, and cutting-edge technology. See you next time!