This copy is for your personal, non-commercial use only. Distribution and use of this material are governed by our Subscriber Agreement and by copyright law. For non-personal use or to order multiple copies, please contact The Epoch Times Reprints.
Despite Record Nvidia Earnings, Stock Analysts Urge Caution on the ‘AI Trade’
‘A company can report outstanding results and still trade at a price that requires perfect performance for many years,’ fund manager Tim Schwarzenberger said.
While many investors took Nvidia’s Nov. 19 announcement of record earnings as validation that the “AI trade” was not simply a bubble waiting to pop but rather grounded in legitimate profitability potential, some insiders continued to sell the company’s shares.
The company reported that it generated $57 billion in revenue in the third quarter, an increase of 62 percent from the third quarter of 2024. The end-of-day announcement, which exceeded many expectations, sparked a brief rally in the company’s shares at the opening on the following morning, but they quickly fell back into the red, falling more than 3 percent in a day and taking market indexes down with them.
Nvidia, the world’s largest company with a market capitalization of around $4.5 trillion, has become a bellwether for valuations of companies tied to artificial intelligence (AI), and although its share price is still up more than 30 percent so far in 2025, there has been significant volatility along the way.
And despite the positive news on earnings, financial analysts say there are reasons to be cautious.
“Nvidia has now beaten expectations for many consecutive quarters, and it has done it again this quarter, but even consistent beats do not settle the valuation debate,” Tim Schwarzenberger, a portfolio manager with Inspire Investing, told The Epoch Times.
“A company can report outstanding results and still trade at a price that requires perfect performance for many years. Earnings describe the present, while valuation reflects the future path that investors are assuming.”
Shares Fall Despite Record Earnings
Since hitting a peak of $207 in late October, Nvidia’s shares have fallen back to current levels of about $180 per share. In step with Nvidia, the tech-based NASDAQ composite index has fallen more than 6 percent from its October peak, and the broader S&P 500 Index, representing the 500 leading publicly traded companies, is down about 4 percent over the same period.
“Nvidia’s earnings make it clear that at least part of the AI story is real—its revenue growth and margins are the kind companies dream about,” Peter Earle, director of economics at the American Institute for Economic Research, told The Epoch Times. “But when one firm reaches multitrillion-dollar status and dominates the entire market, investors are obviously baking in years of uninterrupted AI spending.”
Calling the recent selloff in Nvidia shares “portfolio housekeeping, not a stampede for the exits,” Earle said that “people are trimming their Nvidia-heavy exposure and shifting a bit toward cheaper sectors that might hold up better if tariffs, high costs, and shaky consumer demand slow things down.”
For many investors, the concern is that, despite positive earnings, stock valuations are simply too high relative to the ability of AI technology to generate actual profits and too concentrated in a short list of tech companies. Since the launch of ChatGPT in 2022, AI-affiliated stocks have comprised about 75 percent of all returns in the S&P 500, as well as 80 percent of earnings growth and 90 percent of capital spending growth, according to a Sept. 24 analysis by Michael Cembalest, chair of JP Morgan’s Investment Strategy Group.
In addition, companies involved in information technology or communications comprise more than 40 percent of the S&P 500, according to a September report by U.S. Bank Wealth Management, and just seven tech companies—Alphabet (Google), Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla—now represent about 36 percent of the total market cap of the 500-company index.
One Company Moving Markets
Given that Nvidia had at its height a market capitalization of $5 trillion, the recent decline in just this one company’s share price has wiped out $500 billion in market value in just a few weeks.
“Nvidia has become so large that its earnings influence the entire market,” Scharzenberger said.
“The strong results provide short term relief, but they do not resolve the deeper issue that a very narrow group of companies is driving most of the market’s gains.
“If Nvidia were a country, it would be the second-largest weight in the main global equity index behind only the United States. A single company with that level of influence creates fragility because even a small change in expectations can move large parts of the market.”
The heavy concentration in AI-related companies creates elevated risk, not only for equity markets but for the U.S. economy as well.
“If the AI trade proves overhyped, the fallout could be big simply because so much of the market now depends on a few AI giants,” Earle said.
“A sharp pullback would dent household wealth and could easily spill into spending and hiring, especially for small businesses already nervous about shifting trade policies and rising input costs. Even with the Fed moving to a more accommodating stance, the economy isn’t immune to a confidence shock that large.”
The Circularity of AI Spending
Market analysts also caution that much of Nvidia’s profitability comes from capital expenditures from other tech companies, whose shares are equally inflated. The majority of Nvidia’s fourth-quarter revenue—$51 billion of the $57 billion—comes from the design and development of massive data centers that provide the infrastructure for artificial intelligence to thrive.
“The tremendous amount of investment into AI has parallels to the late 1990s, when telecom companies aggressively ramped up capital expenditures to support the internet boom,” William Flaig, financial adviser and co-founder of the American Conservative Values exchange-traded fund, told The Epoch Times. “By 2001, it became clear that supply had far outpaced demand, and the telecom sector collapsed, contributing to the broader dotcom bust.”
While this overcapacity laid the foundation in the long run for today’s global internet infrastructure, Flaig said, many investors lost money betting on the tech industry along the way.
“AI is for real, but that doesn’t mean the markets are going to go straight up,” he said. “All the companies fighting for dominance can’t all be winners.”
At a time of a historic spending boom in AI infrastructure, market watchers also highlight the risk of what they call “circularity,” or the widespread practice of AI companies investing in each other and buying products from each other, thus generating revenue and providing capital for new investments.
“Some of the spending moves in a loop, with AI labs, cloud providers, and chipmakers investing in one another and then buying from one another,” Schwarzenberger said. “That type of circular activity can lift near-term earnings without proving long-term profitability.”
Insiders Stepping Out
Stock analysts say it’s noteworthy now to see which investors are buying AI shares and which are selling.
“My sense is that there has been some rebalancing among institutional investors, but retail is still fully committed,” Flaig said.
This suggests that market insiders are stepping back from AI, as retail investors, perennial outsiders, step in.
“The positioning of investors raises caution,” Schwarzenberger said.
“Many early investors have been trimming their Nvidia exposure, options data show an increase in downside hedging, and insider selling has remained steady. When early investors start to reduce risk at the same time that the largest allocators are increasing exposure, it usually suggests that the cycle is maturing and expectations are becoming stretched.”
