The Philadelphia Semiconductor Index dropped sharply in early July, capping a run that had seen the index rally roughly 130% over the prior twelve months. Micron fell as much as 13% in a single session. Intel lost 21% of its value over seven trading days. If you only saw the headlines, it looked like the AI trade was unwinding.
What actually happened is narrower than that. This was a valuation correction after an extraordinary run-up, not a sign that AI chip demand itself is slowing. Nvidia still holds somewhere around 70 to 80% of the AI chip market, and the demand story underneath the stock moves hasn’t changed: TSMC reported a 77% surge in net income and record quarterly revenue in the same window, and separately committed another $100 billion to its Arizona operations, bringing its total US commitment to $265 billion. Companies don’t make that kind of capital commitment if they think demand is cooling.
So why the drop? Stocks that rally 130% in a year tend to correct hard on relatively small pieces of bad news, because a lot of future growth is already priced in. A single disappointing earnings detail, a guidance miss, or just profit-taking after a historic run can trigger an outsized move when expectations were stretched that thin. Intel’s decline has a more specific story behind it, tied to its own execution concerns rather than the broader AI chip demand picture, even as the company hit a real technical milestone by becoming the first to ship high-volume chips on ASML’s newest EUV scanners.
The distinction matters if you’re trying to read these headlines as a signal about AI generally. A stock price correcting after a huge run isn’t the same as the technology losing momentum, and conflating the two is how you end up either panic-selling a real trend or dismissing a real warning sign because “it’s just stocks.”