The semiconductor rally and the Magnificent Seven selloff are telling two different stories about the same AI boom — and history suggests one of them is wrong.
The semiconductor rally and the Magnificent Seven selloff are telling two different stories about the same AI boom — and history suggests one of them is wrong.

A widening divergence between surging semiconductor stocks and the Magnificent Seven entering correction territory is raising questions about the sustainability of the AI-driven bull market.
"The divergence can be put down to Magnificent Seven shedding 6% on an equal-weight basis compared to the other 493 stocks in the S&P 500 climbing 0.7%," Stuart Kaiser, head of US equity trading strategy at Citigroup, said.
The Philadelphia Semiconductor Index has surged more than 80% year to date, fueled by AI chip demand from the very tech giants now under pressure. Nvidia, the index's largest component, has seen its shares climb more than 120% in 2026, while Sandisk has soared 780% and Micron Technology has more than tripled. Meanwhile, the Magnificent Seven — Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla — have collectively entered correction territory, with the group falling 6% on an equal-weight basis last week alone, according to Citigroup.
The divergence matters because the semiconductor rally depends on the Magnificent Seven's willingness to keep spending. If the companies writing the checks for AI infrastructure begin to pull back, the chip boom could lose its primary demand driver. "This is a constructive sign that the bull market can remain intact," Sam Stovall, chief investment strategist at CFRA, said, pointing to broadening market leadership beyond semiconductors into financials and health care.
A Rotation, Not a Collapse
Citigroup's analysis of momentum drawdowns over the past 22 years shows that 70% of such events lead to a rotation into laggard sectors rather than a broad market selloff. The 3.5 percentage point outperformance of the S&P 500 equal-weight index over the market-cap weighted version last week was the fourth best out of 1,903 weekly periods since 1990, and the strongest since November 2020.
Kaiser identified two possible scenarios. In a flat-to-higher market, investing in laggards like software or shorting momentum stocks should work. In a de-grossing event, where momentum stocks are sold off and markets fall broadly, puts on the Nasdaq 100 would be the best hedge. The historical data favors the first scenario.
Hedge funds have already begun adjusting. Fund managers cut exposure to the Magnificent Seven in recent weeks, according to Citigroup data, rotating into sectors that have lagged the AI-driven rally — including financials, health care and select software names.
The AI Spending Calculus
The core tension in the market centers on whether the Magnificent Seven's massive capital expenditure on AI infrastructure will generate sufficient returns. Nvidia alone has seen its data center revenue multiply as cloud providers and tech giants race to build out AI compute capacity. But investor skepticism about the return on that spending has grown, contributing to the selloff in the very companies driving chip demand.
The S&P 500's forward price-to-earnings ratio of roughly 21 is "reasonable" given that the first half of 2026 was characterized by an earnings-driven rally rather than multiple expansion, Stovall said. He maintains a year-end S&P 500 target of 7,730, citing 24 all-time highs in the first half and declining oil prices as tailwinds for consumer spending and corporate earnings in the second half.
For semiconductor investors, the risk is that a sustained Magnificent Seven correction eventually translates into lower capital expenditure guidance. TSMC, which manufactures chips for Nvidia, AMD and Apple, has benefited from the AI buildout but faces single-source concentration risk as the primary foundry for advanced nodes. Any pullback in customer orders would ripple through the entire semiconductor supply chain, from equipment makers like ASML to packaging firms like ASE Technology.
Nvidia shares trade at roughly 35 times forward earnings, a premium that reflects the market's expectation that AI chip demand will remain robust. If the Magnificent Seven correction deepens and CapEx plans are revised lower, that multiple could compress. For now, the data suggests a rotation rather than a rout — but the divergence between chip makers and their biggest customers has rarely been this wide.
This article is for informational purposes only and does not constitute investment advice.