AI chips are trading at their cheapest valuation since 2011. The Philadelphia Semiconductor Index (SOX) sits near 10 times two-year forward earnings.
However, prices have fallen far faster than analyst forecasts. If those estimates get cut, chipmaker shareholders may find the discount smaller than it looks.
According to Bank of America, AI chip stocks are now trading at a 22% discount to the broader S&P 500, based on expected earnings. That makes them look relatively cheap compared with the rest of the stock market.
But there is a catch. These valuations assume that AI spending will remain strong. If major technology companies cut their planned AI investments for 2027 by just 20%, chipmakers could see their expected profits fall sharply.
That means stocks that look cheap today could suddenly become expensive if their earnings fail to meet Wall Street’s expectations.
AI chips the cheapest they’ve been in 15 years. Trading at single digit PEs. Just as we give birth to super intelligence. pic.twitter.com/36ShalppUS
— Charles Edwards (@caprioleio) November 10, 2026
Fears of slowing AI infrastructure spending drove the decline.
On Nov. 5 alone, the SOX index lost 12.2% during Wall Street’s crash. That was its worst session since March 2020.
Nvidia (NVDA) closed Monday at $178.67, 26% below its 52-week high, in a slide into bear territory.
Leadership risk adds pressure. CEO Jensen Huang admitted an affair with a former executive of Megaspeed, a Singapore-based Nvidia chip buyer.
US authorities reportedly concluded she was a Chinese intelligence asset, which Beijing denies. Nvidia’s board opened an independent investigation on Nov. 8 while backing Huang.
A Nov. 11 congressional deadline now arrives days before Nvidia reports earnings on Nov. 17.
AI chip stocks have started recovering from last week’s crash, but the broader market trend remains negative.
The Philadelphia Semiconductor Index (SOX) rebounded nearly 11% in two trading sessions, closing Monday at 10,033.93. However, the index is still struggling to break above important price levels.
For investors, three scenarios matter.
The next major test comes on November 17, when Nvidia reports earnings. Strong results could help restore confidence in AI chip stocks, while disappointing forecasts could trigger another selloff.
Bank of America still expects the sector to fall roughly another 10% before reaching a bottom in early 2027. That suggests the recent rebound may be premature, even with valuations at historic lows.