Nvidia sheds 7.3% as AI spending doubts and Texas data center fights bite

The AI trade just took a punch to the ribs. Nvidia shares slid 7.3%, the chipmaker's steepest fall since April 2025, with the damage spread across the five trading days ending November 4, 2026. The stock closed that day at $200.13, down 3.17% from the prior close of $206.69. Market cap still sat around $4.8 trillion — a 7% haircut on a company that size erases more value than most listed companies are worth in total.

The bleeding started early. On October 30, 2026, shares fell 4.22% in a single session and never really recovered their footing. By early November, Nvidia was trading roughly 16–18% below its 52-week high, which landed somewhere in the $239–$243 range.

Three worries are doing the pushing, and none of them are about whether the chips actually work. First, hyperscaler spending: the giant cloud operators that rent out computing power have been posting heavy capital expenditures on AI, and investors are starting to ask whether those budgets can climb forever. Second, data center politics — reports surfaced of proposed data center moratoriums in states including Texas, a linchpin of US data center capacity, according to Crypto Briefing. Third, the midterms, with concerns growing about regulatory hurdles that could follow the election results.

For context, the last time Nvidia fell this hard, Washington was the trigger. On April 16, 2025, shares dropped around 6.9–7% after news of US export curbs on AI chips to China, leaving the stock at roughly $104.49. It now trades near $200. But the two selloffs don't feel alike: April 2025 was one policy hitting one market. This one is diffuse — spending fatigue, local zoning fights and post-election jitters stacked on top of each other.

The next catalyst is already on the calendar. Nvidia reports earnings on November 17, 2026, and investors will be listening hard for commentary on customer demand. Guidance on data center buildout will get the same scrutiny, since proposed moratoriums could stretch the timeline for new capacity and ripple straight back into chip orders.