KL / RESEARCHBack to research
RESEARCH NOTE 012 · EQUITIES / SEMICONDUCTORS / VOLATILITY

When the AI Trade Broke

Why stocks slid on November 5, why chipmakers fell hardest, and why Friday’s bounce didn’t undo the damage.

Silicon wafer reflecting a red streak against a dark Manhattan skyline
AI-generated editorial illustration. Market figures below are from daily closing data.

On Wednesday, the S&P 500 gained 1.58%. On Thursday, it fell 4.36%, and the Nasdaq Composite lost 5.52%. Friday’s rebound was uneven: the Nasdaq rose 1.09%, the S&P rose 0.31%, and the Dow slipped another 0.46%. The story is not that stocks fell every day. It’s that a one-day selloff more than erased the broad market’s post-election rally. [1]

01 / THE TAPEThe rebound was smaller than the drop.

Thursday’s weakness reached beyond megacap tech: the Russell 2000 dropped 4.45%. But the scale of the chip losses tells you where the pressure was most acute. Nvidia fell 9.26% and Micron fell 17.46%. Even after Friday’s bounce, both were still well below Wednesday’s close. [1]

THE DROP AND THE BOUNCEDaily close-to-close price moves · November 5–6, 2026
AssetThu · Nov 5Fri · Nov 6Two-day
S&P 500−4.36%+0.31%−4.05%
Nasdaq Composite−5.52%+1.09%−4.50%
Russell 2000−4.45%+0.89%−3.60%
Nvidia−9.26%+2.78%−6.74%
Micron−17.46%+5.12%−13.23%

Two-day returns compare Friday’s close with Wednesday’s. Nasdaq refers to the Composite, not the Nasdaq-100. [1]

02 / THE REPRICINGThe question shifted from “how big?” to “how soon?”

The catalyst was policy uncertainty. Reuters reported that investors were weighing whether the midterm results could bring more scrutiny to data-center expansion. That was a change in perceived risk, not evidence that a nationwide moratorium had been enacted. The market could mark down the probability of a fast buildout before any project changed its schedule. [2]

The vulnerability was the timing of the payoff. AI projects require spending on chips, power and buildings before the revenue arrives. If construction slips, suppliers may see orders move out and operators may wait longer to earn back their spending. Even if long-term AI demand is intact, a delay can lower what those future cash flows are worth today. That’s one plausible reason the chip selloff was more severe than the index selloff. [2]

HOW A BUILDOUT DELAY CAN REACH STOCK PRICESIllustrative mechanism · not evidence that orders were canceled
  1. 01 · Permitting riskInvestors allow for a later data-center opening.
  2. 02 · Spending shiftsChip deliveries and capacity purchases could move out.
  3. 03 · Payoff moves outCash flows that arrive later are worth less today, all else equal.

03 / THE CROSS-ASSET CHECKVolatility jumped. Long Treasuries barely moved.

The VIX closed at 28.18 on Thursday, up from 16.61 on Wednesday, before easing to 25.64 on Friday. That’s consistent with investors pricing more near-term uncertainty, not a forecast of another down day. Meanwhile, TLT, a long Treasury ETF, rose just 0.21% on Thursday. One ETF isn’t a read on the entire yield curve, but the contrast makes me cautious about explaining the whole move as a rates shock. [1]

MY PERSPECTIVEI’d treat Friday as a bounce, not a verdict. The useful test is whether the selloff changes what companies actually plan to spend. I’d watch hyperscaler capex guidance and chip order timing first. If those hold up and more stocks join the recovery, I’d give more weight to a valuation reset. If spending slips, chip orders soften and weakness spreads, I’d take the earnings risk more seriously.

Sources & research notes

Fixed research snapshot through the November 6, 2026 close. Historical analysis for education, not investment advice.

[1] Yahoo Finance daily closes: S&P 500 · Nasdaq Composite · Dow · Russell 2000 · NVDA · MU · VIX · TLT. Calculations use the archived November 6 site snapshot.

[2] Reuters, November 5: Nasdaq set to confirm correction as AI spending worries mount. Used for reported market context; figures above are final closes, not the intraday quotes in the Reuters report.

Related: Is the Market Falling—or Just Big Tech?