More and more clients are asking Goldman Sachs: Is the US stock market "too optimistic"? What’s next for "AI trading"?

```

“Are we being too optimistic?” This is the question that Goldman Sachs strategists have heard most frequently from clients recently.

Indeed, the market’s concerns are not unfounded. According to Chasewind Trading Desk, Goldman Sachs’ latest report data shows that after a 32% surge in 2024, AI-driven stocks have risen another 17% so far in 2025. With such rapid gains, the S&P 500’s expected price-to-earnings ratio has reached 22, ranking in the 96th percentile of historical records.

However, after in-depth analysis, Goldman Sachs has given a relatively moderate answer: Optimistic, but not yet irrational.

The report points out that the current long-term earnings growth expectation implied by market pricing is about 10%, which is only slightly higher than the historical average of 9%, but far lower than 16% during the tech bubble in 2000 and 13% at the market peak in 2021.

Even among the brightest “star stocks” in the market—the large technology stocks—their valuations remain relatively restrained. Report data show that the average expected P/E ratio for the top five technology giants by market value (Nvidia, Microsoft, Apple, Google, Amazon) is currently 28, significantly lower than the 2021 peak of 40 and the 50 during the tech bubble era. Goldman Sachs believes that compared to the last two market peaks, today’s valuations are expensive but still have some distance to go.

“Halftime Battle” of AI Trades: Infrastructure Frenzy and Growth Bottlenecks

Goldman divides the evolution of AI trades into different stages, and the current market is evidently still immersed in the “second stage”—an infrastructure construction frenzy.

The fuel for this frenzy comes from the hundreds of billions of dollars in capital expenditures by major cloud service providers such as Amazon, Google, Meta, and Microsoft. According to Goldman Sachs report, this year alone the forecast for total 2025 capex by these giants has been revised up $100 billion, reaching $368 billion.

This massive investment directly translates into orders and profits for providers of infrastructure such as semiconductors, power equipment, and technology hardware, pushing up their stock prices.

However, beneath the feast lurks risk. Goldman Sachs warns that the rapid growth in capital expenditure “will inevitably slow down,” which poses a valuation risk for “second stage” stocks. At present, these companies’ stock price gains have far outpaced their recent earnings growth trajectory, reflecting the market’s extremely optimistic expectations for their future growth.

Key Risk: Capex Slowdown Could Trigger 15-20% Pullback

Goldman Sachs explicitly identified the core weakness of the current AI trade: dependence on the tech giants’ capital expenditures. Analysts generally forecast that capex growth will slow markedly in Q4 of 2025 and into 2026. Once the inflection point in growth appears, it could put pressure on related stock valuations.

Although predicting this timing is extremely challenging—since market consensus has repeatedly underestimated the scale of tech giants’ investments—this slowdown trend is “inevitable.”

The firm constructed an extreme stress test scenario: if tech giants’ capex suddenly returns to 2022 levels, that would reduce the expected S&P 500 sales growth in 2026 by about 30%.

Their macro valuation model shows that this shock would not only directly hit short-term earnings, but also seriously harm confidence in long-term AI-driven earnings growth, potentially resulting in a 15%-20% de-rating of the S&P 500’s valuation multiples.

What’s Next? The Market Still Awaits an “Earnings Story”

When the infrastructure construction boom peaks, the next stop for AI trades—the “third stage,” companies realizing revenue growth enabled by AI—can these step up?

Goldman Sachs’ observations reveal market hesitation. The report states that investors show “limited interest” in “third stage” companies, especially in the software sector.

The reason is that the market is grappling with a complex question: For many software and services companies, will AI be a catalyst for growth or a disruptive threat? Goldman Sachs analysts point out that investors are worried AI could disrupt existing pricing models, lower industry entry barriers, and compress the profit margins of established software giants.

Unlike the clear winners in the “second stage,” the “third stage” will see obvious differentiation—with both winners and losers. As a result, investors are especially selective. The report emphasizes that the market may demand to see “real and tangible” earnings impact from AI for these companies before truly embracing them.

As for the longer-term “fourth stage”—AI-driven productivity gains—Goldman Sachs believes this is only just beginning. The report cites data showing that although 58% of S&P 500 companies have mentioned AI during earnings calls, mainly in customer support, programming, and marketing, few companies have been able to quantify the specific contribution of AI to current profits.

~~~~~~~~~~~~~~~~~~~~~~~~

The above wonderful content comes from Chasewind Trading Desk.

For more in-depth interpretation, including real-time insights and frontline research, please join [Chasewind Trading Desk Annual Membership]

Risk Reminder and DisclaimerThe market involves risks, and investment needs caution. This article does not constitute personal investment advice and does not take into account any specific user’s investment objectives, financial condition, or needs. Users should consider whether any opinions, views, or conclusions in this article are suitable for their particular situation. You are responsible for your own investment decisions made on this basis. ```