On Thursday 5 November 2026, two days after the US midterm elections, American technology stocks had their worst day in a year and a half. The Nasdaq Composite fell about 5.5% and closed in correction territory, more than 10% below its 6 October record. The S&P 500 fell about 4.4% and the Dow lost 1,833 points (3.4%).
The losses were concentrated in the companies that make, power and rent out AI computing. Nvidia fell more than 9%, Broadcom more than 10%, and the Magnificent Seven lost more than $1 trillion of market value between them in one session.
The short version
- Nasdaq: down about 5.5% to roughly 24,315, its worst day since 4 April 2025.
- S&P 500: down about 4.4% to 7,353.73. Dow: down 1,833 points to 51,523.06.
- Tech: the S&P 500's technology sector fell 7.3%, its worst day since March 2020. The Philadelphia semiconductor index fell 12.2%.
- Nvidia: down about 9.3%, losing more than $440 billion in value and turning negative for 2026.
- Why: the midterm results raised expectations of tougher rules on AI and on data-centre construction, on top of existing doubts about the return on AI spending.
- Spillover: on Friday the Nikkei fell about 7% at the open and Korea's Kospi fell more than 8%, triggering a trading halt.
What happened on 5 November 2026?
| Index or stock | Move on 5 November |
|---|---|
| Nasdaq Composite | about -5.5% |
| S&P 500 | about -4.4% |
| Dow Jones | -1,833 points (-3.4%) |
| S&P 500 technology sector | -7.3% |
| Philadelphia Semiconductor Index | -12.2% |
| Nvidia | about -9.3% |
| Broadcom | more than -10% |
| AMD | about -17% |
Every one of the 11 S&P 500 sectors finished lower. On the Nasdaq, losers outnumbered gainers by almost five to one. The VIX volatility index jumped 11.57 points, its biggest one-day rise since April 2025.
Why did AI stocks fall so hard?
The midterms. Reporting by NPR, Axios and others tied the sell-off to the election result, and specifically to the expectation that the new Congress will bring more oversight of AI and of the giant data centres being built to run it. Local opposition to data centres, over electricity prices, water and land use, had been building all year, and it showed up at the ballot box.
Valuations were stretched. By November most of the big AI names were priced for years of uninterrupted growth in AI spending. When a stock is priced like that, a hint that the build-out might slow is enough to move it 10% in a day.
Doubts about returns were already there. Investors had spent weeks asking when the hundreds of billions going into AI infrastructure would turn into profit. AMD had just reported record quarterly revenue and its shares still fell about 17%, which tells you the market was no longer rewarding growth alone.
Speed. When a crowded trade unwinds, funds that use the same models sell the same stocks at the same time. That turns a bad day into a historic one.
How did markets in Asia and the Gulf react?
Asian markets opened on Friday 6 November into the US sell-off. Japan's Nikkei fell about 7% shortly after the open before recovering part of the loss. South Korea's Kospi dropped more than 8%, triggering a 20-minute circuit breaker, and Taiwan's main index also fell more than 8%. The heaviest selling was in the chipmakers that supply the AI build-out: Samsung Electronics, SK Hynix and TSMC.
"The theme we are watching most closely is AI scrutiny," Vijay Valecha, chief investment officer at Century Financial in the UAE, told The National.
Pakistan went the other way. The KSE-100 had its best week in six months, because the same week brought a sharp fall in oil, and Pakistan's market is driven by energy, banks and cement rather than by AI.
Is this the AI bubble bursting?
One day does not answer that, and we would be wary of anyone who says it does. A 10% fall from a record is a correction, not a crash. Corrections have happened inside every long bull market.
What has changed is the question investors are asking. For three years the question was how much companies would spend on AI. Now it is whether they will be allowed to build it as fast as planned, and what it will earn when they do.
What does it mean for businesses using AI?
Very little, in the short term. A falling Nvidia share price does not make the tools you already use stop working. Three second-order effects are worth knowing about:
- Pricing pressure may ease, or tighten. If data-centre building slows, AI compute could get scarcer and dearer. If AI companies need to show profits faster, they may raise prices. Either way, read your contracts; we covered the clauses that matter in our guide to AI vendor risk and lock-in.
- Electricity is now political. Data-centre power demand is a large part of why this became an election issue, as we explained in what AI data centres are doing to US electricity costs.
- Regulation is coming either way. If you sell into the US, the patchwork of state AI and privacy rules is already the practical issue, whatever Congress does next year. Start with our operator's guide to US state AI rules.
This post describes what moved and why. It is not investment advice.
Frequently asked questions
Why did the Nasdaq crash on 5 November 2026?
Investors sold AI and chip stocks after the US midterm elections raised expectations of more oversight of AI and of data-centre construction, on top of concerns about high valuations and the return on AI spending.
How much did Nvidia fall?
About 9.3% on 5 November, losing more than $440 billion in market value and turning negative for the year.
Is the Nasdaq in a correction?
Yes. A correction is a fall of 10% or more from a record high. The 5 November close put the Nasdaq Composite about 12% below its 6 October record.
How did Asian markets react to the US sell-off?
On 6 November the Nikkei fell about 7% at the open, the Kospi fell more than 8% and triggered a trading halt, and Taiwan's index fell more than 8%.
Sources
CNBC for sector, breadth and Nvidia figures; The National for Asian markets and the Century Financial quote; Axios; NPR; Fox Business; index closes from Associated Press via TechStartups. As of 9 November 2026.
