Don't believe in the AI bubble theory too early! Chen Guo: The biggest "unexpected" in 2026 is all in the Chinese market

Q1: After the volatility of the AI leaders in the US stock market, the market is asking: Has AI risen too much? What is your view on whether AI has entered a bubble?

Chen Guo: I have always believed that "no bubble, no bull market." If the market is always the most rational static pricing, it is actually difficult to have a real bull market. The key is not whether there is a bubble, but rather—when will the bubble burst.

I have a formula to judge whether a technology bull market bubble will burst, which roughly includes three conditions:

First, valuations must be "seriously high." Currently, the representative indices like the magnificent seven, NASDAQ, and S&P have relatively restrained PEs, around 20 to 30 times. In comparison, during the internet bubble burst, many companies had valuations of 100 times or 200 times; Cisco at that time was somewhat like today's NVIDIA; Amazon was more like today's OpenAI—back then it wasn't even profitable, so it couldn't even be calculated in terms of PE. Therefore, if valuations are only moderately high, it is difficult to form a bubble that bursts.

Second, liquidity must continue to tighten. This includes sustained interest rate hikes or a significant tightening of micro liquidity. For example, the downturn cycles of the US stock market in 2000 and 2007, or our A-shares in 2015, had a core issue of tightening liquidity combined with high valuations.

Third, technological iteration slows down, or revenue growth shows "foreseeable peaks." This does not refer to this year's growth peaking, but rather that the market can see that next year or the year after will peak, making it naturally difficult for valuations to rise further.

Currently, none of these three conditions are met. Therefore, I believe that while "bubble formation" can be discussed, "bubble bursting" is far from being the case.

If you use the most stringent valuation system, you can certainly say "this is a bubble," for example, OpenAI is not profitable, and cannot even be discussed in terms of PE. But I still say: no bubble, no bull market. From the current perspective of valuations, liquidity, and technological cycles, there are no conditions for a bubble to burst.

Q2: With changes in overseas fiscal stimulus, technology regulation, and interest rate paths, how will these external risks transmit to the domestic market?**

Chen Guo: As domestic market investors, we subjectively hope for an "independent market," unaffected by overseas influences. However, from the actual situation, as of now, the correlation between A-shares, especially in technology, Hong Kong stocks, and US stocks is still very strong, and changing this paradigm will take time.

Logically, the core driving force behind the current rise in A-shares is still technology and internationalization. The technology sector is currently in a highly concentrated global pattern, and in many aspects of AI, the US remains the global industry leader. Therefore, it has a "reference system" in industrial judgment and valuation pricing, which will affect us. If some actions by the Federal Reserve or changes in liquidity affect the valuations of US tech stocks, we will be indirectly impacted In the past, we have actually enjoyed this benefit: high overseas valuations, which we can benchmark against. However, we cannot say that when it corrects, we are completely unaffected. The reality is often that A-shares do not benefit enough, and the reaction during a pullback is more pronounced. Therefore, we need to assess more seriously: what level are the valuations of Chinese-related industry companies at, and how strong is the market's recognition of them? If a company's intrinsic value is strong enough, its valuation has a safety margin, and market confidence is also strong enough, then theoretically, fluctuations in overseas liquidity and valuations will not have such a significant impact on us. However, currently, the overall confidence in the market is not that strong.

Another important point is that there are many companies in A-shares with a relatively high proportion of overseas revenue. Looking at the major indices, the overseas revenue proportion is roughly close to 20%, and the profit proportion is even higher, with some industries having profit proportions close to or exceeding half. In this case, if the U.S. AI industry declines, dragging down the U.S. economy and the global economy, it will actually affect these Chinese companies—even if they are not AI companies themselves, as long as they are in an internationalized environment, they will be impacted.

Therefore, Chinese investors in this bull market must possess a global perspective to assess: where the global industry stands and what cycle the global bull market is in. Of course, we also hope that in certain AI fields, China can lead in the future, at which point pricing will not be entirely behind the U.S. But I believe this process may only gradually unfold after 2026.

Q3: In your report, you mentioned the "three major expectation differences" for judging next year's market. Can you quickly summarize them for everyone?

Chen Guo: Since the bottom last year, we have actually been strategically optimistic. The market has risen from over 2,600 points to around 4,000 points this year, which is a significant increase, not to mention some technology indices. Therefore, the market naturally asks: have so many favorable factors already been priced in? What will drive it higher in the future?

My view is: there are at least three directions that may perform better than the current mainstream market expectations.

The first expectation difference is corporate earnings. From 2022 to 2024, the overall earnings of listed companies have been significantly below expectations, which is one of the main reasons for that bear market. The corporate earnings corresponding to the major indices are basically in negative growth, even though GDP is around 5%, earnings are negative.

This year is different: as of now, corporate earnings are in positive growth and have not fallen below expectations. The market's expectations for earnings next year are not high, and I believe they may exceed expectations. The core reason is not simply a recovery in total demand, but rather changes in industry supply and demand relationships: the structure of most industries is improving, capital expenditures of listed companies are significantly declining, even showing negative growth, the growth rate of construction projects is close to 0, and fixed asset investment has been significantly reduced. As long as revenue maintains positive growth and supply contracts, leading companies will benefit, and ROE will improve.

If we add some policies that "counter involution" and optimize the supply side, it will be even more beneficial. But even without particularly strong administrative push, many companies are also making supply-side contractions and improving quality and efficiency, which will be reflected in the earnings in the coming years The second expected difference is the dividends of Chinese enterprises in internationalization. Next year, the dividends for Chinese enterprises in globalization may be more evident than this year. Major global economies are generally in a combination of "loose monetary + loose fiscal" policies. The U.S. economy appears to be doing well this year, but largely relies on AI capital expenditures for support, while the other half of the economy is actually quite weak. If there are continued interest rate cuts and fiscal deficit support, there is still room for recovery.

In such an environment, many Chinese companies are still increasing their market share overseas. The trade war that everyone is worried about has not reversed this trend so far. Even if Trump 2.0 raises some tariffs, the tariff levels on China are still differentiated and not a comprehensive suppression compared to some allied countries. Overall, China's competitive advantage in global trade is strengthening, which will ultimately be reflected in the profits of listed companies.

The third expected difference is China's competitiveness in AI and the larger geopolitical environment. The market is very concerned that China will lag behind in AI in 2024; by 2025, there will be growing confidence that "we can keep up"; I believe that by 2026, there may be increasing evidence that China is leading in many areas.

Currently, if we only look at large model scores, some platforms may still have the highest scores from U.S. models, but China's large models are very efficient—achieving comparable scores to leading U.S. models at a fraction of the cost, even one-hundredth of the cost. In the future, it will not only be about the large models themselves but also about entering AI agents and various application scenarios, where China's cost efficiency advantage will be reflected in the commercialization phase.

In the end, business competition is about economics rather than just who has a slight technological edge. Whether it's edge AI (AI smartphones, AI glasses, robots) or the underlying infrastructure like electricity, our advantages in cost and supply are not something the U.S. can easily catch up with in the short to medium term. Therefore, the cost-effectiveness of Chinese solutions in AI application commercialization will become increasingly prominent, and this advantage will gradually be re-evaluated in next year's market.

These factors can currently only be vaguely perceived in the market, making precise pricing difficult—how much will profits exceed expectations? To what extent will AI lead? In which areas will China and the U.S. expand cooperation? Thus, they constitute a clearly defined but quantitatively difficult "expected difference combination" that needs to be gradually realized by 2026, and the market will continuously re-evaluate and price during this realization process.

Q4: Among the three major expected differences, which one do you think is currently the most underpriced by the market?

Chen Guo: I believe the most underpriced aspect is the potential for China to lead in AI.

This actually goes back to the question we started with—the current paradigm of the entire market still assumes that the U.S. is absolutely leading in AI. The U.S. is worried about an AI bubble, concerns about the return on investment not being proportional, fears of excessive capital expenditures, and that commercial applications are lagging behind. Our stock prices will also decline because we default to seeing ourselves as "followers." But if the market gradually sees in the future that China's economic efficiency in many AI applications is better and commercialization is faster, the perspective will be completely different. It may happen that the capital expenditure on AI in the U.S. cannot be recouped in the end, but the economic viability of AI applications in China is good and can be accounted for; the profitability in the U.S. is released very slowly, while the profitability of AI applications in China appears earlier.

From the current paradigm, the pricing in this area is obviously insufficient, and its expectation gap may be very large. Corporate profits may also exceed expectations, but this expectation gap may not be large enough to change the entire narrative; however, if the advantages of China's AI applications are truly recognized and reassessed by the market, it could change the "main narrative" in this round of bull market between China and the U.S. Therefore, I believe this is the most important and easily overlooked expectation gap.

Q5: What changes in liquidity do you think will be most noteworthy in 2026?

Chen Guo: I think the core still lies in micro liquidity, that is, whether incremental funds are genuinely flowing into the stock market.

This year, there has been some inflow of incremental funds, but it is not particularly obvious. The reason is that this is the first year of the bull market, which is essentially unfolding slowly amid skepticism. From an asset allocation perspective, the shift in people's mindset takes time. Historically, incremental funds often become more apparent in the second year of a bull market.

Additionally, China's funding structure is layered, and it is necessary to distinguish between high-net-worth individuals and the middle class. This year, high-net-worth individuals have entered the market, but they are more likely to do so through bank wealth management, fixed income +, insurance, and other non-bank financial institutions, rather than directly buying stocks. The middle class, which was very enthusiastic about public funds from 2019 to 2021, is now generally hesitant, with many still in a "break-even mentality": once they break even, they redeem, so currently, public funds are still in a net redemption state.

If the market moves relatively steadily in the future, with periodic profit-making effects and continuously "unexpected" events consolidating confidence, then the first step would be for public funds to shift from net redemption to net subscription. If we look at public fund-related indices, they are still about 15% away from their historical peak; if there is another wave of new highs, it will no longer bear the pressure of "break-even redemption," leading to real net inflows.

For high-net-worth individuals, they do not necessarily pursue extremely high returns but pay more attention to the cost-effectiveness of returns and volatility, that is, the Sharpe ratio. This year, from the perspective of major global asset classes, the Sharpe ratio of the Shanghai Composite Index ranks very high: the increase is not the highest, but the volatility is very low and stable. This is different from the short-term market in 2024. If we comprehensively assess returns and volatility, the Shanghai Composite Index is a better choice compared to gold, bonds, and many overseas indices. This will attract high-net-worth individuals to further increase their allocation to equity assets.

Another critical time point is: a large amount of time deposits will mature in the first quarter of next year. After maturity, when renewing, the interest rates will be much lower than before. This will force funds to seek new asset allocation outlets, and from this perspective, there is a basis for incremental funds flowing into the stock market Finally, there is a "bonus," which is foreign capital. Since 2022, there has been no significant net inflow of foreign capital overall. I believe that by 2026, there is a considerable probability that net inflows will resume. On one hand, profits are moving from "below expectations" to "close to expectations," and then to "expected to exceed expectations"; on the other hand, Sino-U.S. relations are relatively stable, and in terms of global allocation of AI, it should naturally involve both China and the U.S. Additionally, considering China's potential advantages in AI applications, theoretically, foreign capital has the motivation to increase its allocation to Chinese equity assets. Although this area has relatively greater uncertainty, overall, I believe that the inflow of incremental funds next year is likely to bring real "surprises."

Q6: Among growth, cycles, consumption, and other main lines, which sectors are most worth closely tracking?

Chen Guo: First, I agree with your premise—there will not only be one main line; there are opportunities in many directions.

From the perspective of funds, incremental funds have their own "aesthetic." The style of sectors and industry gains are actually highly correlated with the preferences of incremental funds. In the past, the structure of incremental funds was relatively singular, but next year, there may be multiple layers of funds entering the market together, including foreign capital, high-net-worth individuals, the middle class, private equity, public funds, and proprietary trading, which will naturally lead to a more "diversified" market style. In this case, we need to look at both fundamentals and valuations.

From a fundamental perspective, I would roughly summarize next year's layout into three letters: A, B, C.

A is AI. AI remains the strongest industrial clue, but the "focus" of sectors and individual stocks is changing. In the past, the market's focus was on computing power, such as NVIDIA, but it is now gradually shifting towards the application layer, such as Google, and will continue to evolve in the future. Overall, the focus will spread more from computing power to the application end. The real "AI bull stocks" may not necessarily be the technology companies we traditionally recognize; they could very well be traditional enterprises that efficiently use AI to significantly enhance profit margins and market share. Such companies may also become core assets in this round.

B is Biotech. In the context of a rate-cutting cycle and the acceleration of China's internationalization, besides AI, I believe Biotech is also a line worth paying attention to. Globally, the only companies that can achieve a market value of trillions of dollars, besides technology companies, are large pharmaceutical companies. China's gap in this area is narrowing, and the potential for imagination is considerable. Coupled with AI's enhancement of efficiency in pharmaceutical research and biotechnology, China's advantages in cost, cost-effectiveness, and efficiency will be very evident, so I believe this area is also a focus.

C is Commodity. On one hand, AI itself requires a large amount of infrastructure investment; on the other hand, the world is stimulating the economy and loosening liquidity, while in the past few years, the supply of many commodities, especially some "new minor metals," has been tight. Under this combination of "tight supply + monetary easing + new demand," there will also be many opportunities in the commodity-related sectors Of course, in actual operations, the market next year is likely to be a highly rotational and strongly differentiated environment. You cannot rely solely on subjective imagination about who is stronger; it is necessary to combine a large amount of fundamental data and quantitative data for dynamic adjustments. Internally, we will also use some industry rotation models; it may be that the first quarter is style A, the second quarter is style B, and the third quarter rotates back to some old style, which is highly probable.

If you do not have much energy to grasp industry rotation, the simplest way is to: select a portion of indices or representative companies from lines A, B, and C, and create a structured allocation. This is a thought process I believe can be considered for next year.

Risk Warning: The master class is a platform for selecting personnel from third-party compliant institutions to teach investment research theory courses. The content taught does not constitute buy or sell recommendations for any specific products or investment advice. The opinions expressed in the platform courses are for learning and reference only and do not represent the opinions or views of Wallstreetcn, nor do they address the user's specific investment goals, financial situation, or needs. The market is volatile and uncertain, and the platform is not responsible for any losses incurred by you relying on the opinions or information from the courses. Investment carries risks; please make decisions cautiously