The S&P 500 fell more than 1% in a single day, yet over 300 of its components closed higher? Wall Street analysts believe that under the AI boom, broad market indices are becoming increasingly poor indicators of the true performance of "typical stocks."
The U.S. stock market is experiencing a rare internal divergence: while AI and semiconductor stocks, which drove the four-year bull market, are rapidly pulling back, consumer staples, small- and mid-cap stocks, and a broad range of traditional industries are gaining strength. Capital is not simply exiting the market but is rapidly shifting between these two camps, causing major indices and most individual stocks to move increasingly out of sync.
On Thursday, the S&P 500 fell 1.2%, posting its worst single-day performance since July, while the Nasdaq Composite dropped sharply by 2.6%; however, more than 300 components of the S&P 500 actually closed higher. The Dow Jones and Russell 2000 indices even edged slightly higher.
This phenomenon is uncommon. According to Dow Jones market data, the last time a similar extreme divergence occurred—where the S&P 500 declined significantly while most of its components rose—was in January 2025.
AI stocks have fallen so sharply that they outweigh the gains of most other stocks.
The core reason for this unusual movement is that U.S. stock indices have become highly dependent on a small number of large technology and AI companies.
The top 10 companies by market capitalization in the S&P 500 currently account for approximately 40% of the index's total market value. Because the S&P 500 is market-capitalization weighted, a sharp decline in just a few large stocks is enough to drag down the entire index, even if most other companies' stock prices are rising.
This structure was particularly evident on Thursday. The Philadelphia Semiconductor Index plunged more than 6% in a single day and has now declined over 26% from its June high, entering a technical bear market. Just months ago, the index had doubled rapidly amid heavy inflows into AI chips and computing power “bottleneck” assets.
Meta has become one of the triggers for new concerns. Although its earnings report exceeded market expectations, the company experienced its first cash outflow in over a decade due to AI-related capital expenditures. Simultaneously, it announced plans to raise additional funds through a new share issuance to continue investing in AI, causing its stock price to drop by approximately 7%. The market is now reassessing: how much external financing will massive AI investments require, and when will these investments generate sufficient returns?
In contrast, traditional defensive sectors such as consumer staples showed significant strength. Stocks like General Mills and Campbell’s rose sharply that day, but their gains were insufficient to offset the drag from large AI-weighted stocks on the index.
Falling indices and rising common stocks are becoming the norm in 2026.
More importantly, this is not an isolated daily anomaly. Nationwide’s chief market strategist, Mark Hackett, describes the 2026 U.S. stock market as increasingly resembling a seesaw: one side consists of AI stocks, while the other comprises nearly everything else—when one side rises, the other often falls.
As of Thursday, the equal-weight S&P 500 is still on track to rise this week, while the Philadelphia Semiconductor Index, which represents core AI trading, continues to decline. If this pattern holds through Friday, the two indices will have moved in opposite directions for eight consecutive weeks—the longest such streak on record since data collection began in the mid-1990s.
BTIG strategist Jonathan Krinsky found that in over 30% of trading days this year, the S&P 500 moved in the opposite direction of most of its components—a ratio higher than in any full year since 1996.
This means it is increasingly difficult to assess the overall state of the U.S. stock market by observing the S&P 500 index alone. The index may decline while most individual stocks rise, or it may rise due to a rebound from a few AI giants, while a large number of other stocks in the market show weak performance.
The funds are not exiting, but rapidly moving between the two camps.
This split also explains the intense “back-and-forth” sentiment recently felt by investors. Farzin Azarm, Managing Director of Equity Trading at Mizuho Americas, believes that capital is currently being reallocated within the market rather than fully exiting risk assets.
When AI stocks were sold off, capital flowed into consumer staples, small- and mid-cap stocks, and U.S. Treasuries; when chip and data center stocks rebounded, capital quickly returned to growth sectors.
A similar pattern occurred last week. On Monday, after eight consecutive trading days of declines and a cumulative drawdown of approximately 15.7%, the Philadelphia Semiconductor Index rebounded by 1.6% in a single day, prompting investors to re-enter chip and AI-related stocks. Just days later, renewed concerns over AI funding and capital expenditures triggered another wave of large-scale selling.
High-frequency rotation is accompanied by institutions reducing risk. Data from Goldman Sachs Prime Brokerage shows that hedge fund clients have recently significantly reduced their overall leverage exposure. Some traders believe this indicates investors are actively unwinding crowded positions rather than experiencing a broad liquidity crisis.
However, another perspective holds that such extreme divergence cannot continue indefinitely. Krinsky warned that if the pattern of constantly rotating between AI and non-AI stocks eventually breaks down, the market may no longer see “one side falling while the other rises,” but instead shift toward more correlated, synchronized declines.
Thursday’s market action thus revealed a growing structural feature of the current U.S. stock market: although this bull market is still represented by the S&P 500 index, the index has effectively split into two markets with vastly different performances. As long as AI giants continue to account for nearly 40% of the index’s weight, their price movements may still determine the index’s direction, even as the majority of U.S. stocks follow a different path.