Bridgewater Ray Dalio: When the market is dominated by AI technology stocks, how should you invest?

Bridgewater Associates founder Ray Dalio points out that the market is highly concentrated in AI tech stocks, and both history and mathematics prove that diversification is superior to concentrated bets; he also warns that the real return of U.S. stocks over the next 5–10 years could be between -5% and -10%.
(Background summary: Bridgewater founder Ray Dalio warns: signs of bubble in the AI market, profitability is the key to collapse)
(Additional context: Ray Dalio warns of the collapse of the three major orders, and the U.S. is entering a long-term downward phase)

Table of Contents

Toggle

• How to respond to the current market situation
• Risks are indeed extremely high
  • *Achieve true diversification
• Expected returns seem relatively low
• Questions raised by the research team
  • My answers

Hedge fund founder and investment master Ray Dalio posted a lengthy article on X platform last night (15th). He shares what he believes is the best investment strategy when the market is highly concentrated in AI tech stocks.

Imagine you are playing bridge, Texas Hold'em, background games, or chess, needing to decide your next move; and you have a computer assisting you in evaluating the situation and suggesting moves. That’s how I approach the investment game. Whether or not you have computer assistance, I believe you should:

Based on the current board setup (i.e., market conditions and influencing factors), think about what actions you should take.

I’ve been involved in the investment game for a long time. My goal now is to pass on my approach—and further, to build a platform that allows all kinds of people to explore the investment game, learn, review their past performance, and prepare for investing. I believe that when facing your cards, there are right and wrong ways to respond. Therefore, when you encounter a certain market configuration, you should ask yourself: “In this situation, how should I bet?” and be able to give a good answer.

Now, I want to share my views on the current market features and the strategies I believe (and am) implementing.

How to respond to the current market situation

What is the most important environmental factor right now? How should we bet in response to these factors?

In my view—and I believe most others share this sentiment—we are currently in an environment dominated by a very small number of companies that lead market trends, characterized by cutting-edge new technologies (mainly AI). These companies account for a significant portion of market value and have a huge impact on the economy and markets. As with previous similar situations, the new tech sector is filled with excitement, uncertainty, and volatility, which then propagates to global stock markets. Therefore, the turbulence and uncertainty in this sector are critically important.

Additionally, there are other major uncertainties stemming from what I call the “Five Major Forces”: 1) debt and monetary dynamics; 2) political and social issues that could significantly impact taxes (and other political factors affecting markets); 3) geopolitical influences (e.g., wars); 4) natural events; 5) development of new technologies. I input these scenarios into my investment system, thinking about how to position myself in this context, while also making my own active judgments.

When considering how to bet in response to these situations, the core question is: do you want to: a) further overweight new tech beyond the implied weights in broad stock indices (like S&P 500), overweight this emerging sector or your top-quality few stocks; b) maintain roughly index-level exposure; or c) diversify away from this concentration risk?

Almost everyone hopes to buy the best investment targets and strives for that. Currently, a new technology seems to be changing everything. However, history shows that during this phase of the tech cycle, most people fail because they bet too heavily on a few leading tech stocks. This makes logical sense and has been repeatedly validated. Although this time’s AI technology is unique, history has seen many other similarly regarded “unique” new technologies that can serve as references. We should examine these precedents; if we choose to ignore them, we need strong reasons to say “this time is truly different.”

Risks are indeed extremely high

Every great wave of new technology in the past has played out with similar logic. High risk and high uncertainty are inherent features of these new tech companies. Looking back at their performance under such configurations, even top revolutionary tech giants (like Microsoft and Apple) have experienced severe setbacks.

Moreover, when these new tech companies emerged (not in hindsight, but at the time), it was difficult to judge which would succeed and which would fail (e.g., IBM). Reviewing these cases, it’s clear that outstanding new tech companies inherently have highly uncertain futures. For example, they tend to overinvest or underinvest—because underinvestment guarantees failure, but they cannot precisely predict the future to know if they are overinvesting. Both over- and under-investment carry high costs.

Additionally, they cannot accurately foresee all external influences—such as monetary tightening, wars, major tax reforms—that impact them. As a result, they go through dramatic cycles of rises and falls: exciting investors initially, then scaring out the less committed, causing exaggerated market swings. More importantly, just as these new technologies and companies disrupt the old order, most will eventually be overtaken by newer technologies and companies in unpredictable ways. We should consider that this risk could also be present in current tech stocks. The potential impact of quantum computing is one such “known unknown,” but what about risks we haven’t yet imagined?

What about risks from competitors? Take China, for example. China is producing and promoting AI technology, but its policymakers’ views on the economy and AI differ greatly. We are in a tech war, with each country’s leaders believing they must win.

They have different understandings of AI and its impact on the economy and public welfare, and tend to offer this technology for free or at low cost—because it can bring enormous productivity gains and improve overall living standards. They see societal benefits as more important than corporate profits. I believe they will compete on the international stage, just as they have in automobiles, solar panels, batteries, and many other products.

This situation is very similar to many historical cases. I can’t help but think of how, during the decline of the Dutch Empire and the rise of the British Empire, Britain defeated the Dutch in shipbuilding and other key industries. Also, geopolitical tensions around Taiwan may lead us to consider a strategic possibility: China might use blocking chip exports to Taiwan as a geopolitical tool. AI stocks face other risks, such as rising wealth taxes and other tax increases—forcing holders of large positions to sell; or rising anti-AI sentiment, which could limit corporate development.

I could list many more concerning factors, and an equally long list of huge opportunities that AI might bring—opportunities I want to bet on. I am not trying to predict how these risks will evolve, nor am I saying you shouldn’t invest in AI companies.

I am simply saying: the market has significant concentration risk, which is undeniable; and in the face of such market configurations, you should know how to respond. Based on my study of all analogous cases and logical reasoning, I am confident that the risks are high, and the best way to respond to this market setup is:

*Achieve true diversification

You may know that my investment credo is diversification, and my “investment holy grail” is striving to hold 15 high-quality, uncorrelated, risk-balanced investments. In other words:

A well-constructed, high-quality, diversified portfolio will outperform concentrated bets (i.e., at the same risk, achieving a higher reward-to-risk ratio through engineering); and the more concentrated the market risk in a particular sector, the more you should diversify—especially when the market is driven by a fundamentally uncertain revolutionary new technology.

This is not opinion; it’s mathematics. For example, if an investment’s reward-to-risk ratio is 0.3 (assuming a 6% return and 18% standard deviation, typical for stocks), holding 5, 10, or 15 uncorrelated investments reduces risk (measured by standard deviation) to 8%, 6%, and 5%, respectively, at the same 6% return.

Thus, holding 15 high-quality uncorrelated investments can increase the reward-to-risk ratio by 4.3 times (from 0.3 to 1.29). If you wish, you can leverage this to achieve higher returns at the same risk level. That’s the fact.

I am fully confident—based on my backtests, my actual returns over a 50+ year investment career, and logical reasoning—that excellent diversification combined with target volatility will, in the long run, outperform most investors’ tendency toward concentrated bets. More specifically, full diversification can produce better risk-adjusted returns than any concentrated bet; and by adjusting volatility to your ideal level, you can achieve higher returns at your desired risk level in any process.

Because I am sharing this approach, it has become my “not-so-secret” secret to investment success. However, I rarely encounter investors who think this way—meaning very few consider the question of portfolio construction (i.e., how a well-structured, diversified bet set compares to concentrated holdings in a single transformative industry’s stocks).

Most people only think about whether those stocks or industries can go up, and how to bet on them. There is a huge performance gap between those who think about portfolio construction and those who don’t. I will share more detailed ideas on how to do this properly at another time.

For all the reasons above, when considering how to play your hand in this market configuration, we should ask ourselves: how large should the concentrated bet be? And then diversify.

Expected returns seem relatively low

High risk is undeniable, and I now want to share a possibly mistaken view: expected returns are low. This outlook on future returns is based on my valuation analysis and bubble indicator readings: real stock returns over the next 5–10 years seem to be around -5% to -10%, though these figures carry considerable uncertainty.

In my view, these stocks are high-risk, long-duration assets—because reliably predicting the distant future is extremely difficult, and they appear overvalued, held by relatively fragile investors.

Questions raised by the research team

In a recent meeting, one of my research team members asked me: why do you think the current market configuration is wrong? How do you know that the lack of diversification isn’t justified—perhaps because some investors believe AI stocks will have very high expected returns; or because when a sector accounts for such a high market cap, the high concentration in indices is natural; or because when the market is enthusiastic about a sector, many investors buy these stocks heavily without making wise, reliable profit forecasts and valuation calculations?

My answer

Stock prices rise for many reasons, and not all are justified. Some investors carefully evaluate prices and believe fundamentals are attractive, pushing prices higher; some hold these stocks because they recognize them as great new technologies and see the price rise as confirmation of their quality; others hold index positions, passively gaining large weights in these stocks.

In my view, you can spend time analyzing these issues and trying to decide what to do; or you can realize that you don’t need to get caught up in this—because you simply lack enough confidence to make a confident bet. You can fully say: “I don’t have enough information to make a bet.” And then, don’t bet.

The trouble comes from thinking you must form a view and believe that view has enough value to justify betting—when in fact, you probably cannot form a sufficiently reliable judgment to support a bet. (Note: I am not suggesting avoiding betting altogether—since you can’t completely abstain, as your money must go into some investment or cash; and most believe cash is the least risky, but long-term, cash is definitely the worst investment. My advice is: even if you have no tactical view on any market, you should know how to diversify your bets fully. The way to do this is to hold a balanced strategic asset allocation as your baseline when you lack strong tactical convictions. But that’s another topic.)

Therefore, I believe that knowing what you don’t know—and thus when not to bet—is just as important as knowing what you do know to make the right bets.

To put it simply, I follow this principle:

Since it’s generally hard to have enough information to justify concentrated bets, the best approach is: only combine your most confident, uncorrelated bets into a diversified portfolio, and adjust overall risk to your ideal level. That’s what I call the “investment Holy Grail.”

At this moment, in response to the current market configuration, I believe no one has enough confidence to make large, concentrated bets on this tech-driven market. In my view, avoiding concentration and achieving diversification is the best way to handle this “not knowing.” I realize this may contradict textbook theories—those that assume markets are efficient and you should “trust the market.”

In summary, the current market’s unusual concentration around a revolutionary new technology should remind us: do not confuse the excitement about new tech with the investment appeal of stocks in that sector, and do not boldly hold large, high-risk, highly correlated concentrated bets—especially when we can, through wise diversification, achieve similar attractive returns at lower risk.

P.S.: While I won’t share my holdings or tactical views to avoid becoming your investment advisor, I will soon share some key perspectives behind these decisions, including my bubble indicator readings and the reasoning behind them.