The technology selloff of November 5, 2026 raises an uncomfortable question: is this a buying opportunity, or the first stage of a prolonged bear market? My assessment is that a valuation correction is more likely than a sustained collapse, but there is no compelling evidence yet that the bottom is in.
The crucial distinction is between the growth of artificial intelligence and the investment returns earned by its suppliers. AI can succeed commercially while semiconductor shares fall because investors previously assumed too much growth, too much pricing power or too little financing risk. Nvidia, AMD and Micron could continue increasing sales and still disappoint market expectations.
The intensity of the latest selloff makes caution appropriate. November 5 reporting described the semiconductor index falling as much as 11%, its sharpest drop since March 2020, and the Nasdaq losing as much as 5% intraday. AMD, Micron, Dell and Marvell were among major technology names experiencing double-digit declines. These are intraday observations, not closing figures.
One plausible outcome is a recovery after several weeks or months of turbulence. In this scenario, AI investment plans remain broadly intact, earnings estimates stabilize and investors gradually regain confidence. A further decline before recovery would not be surprising. I assign this outcome a subjective 45% probability.
A second scenario is a deeper technology bear market, with the Nasdaq ultimately falling 20% to 30% from its October high. This would become more likely if hyperscalers reduce capital expenditure, data-center construction is delayed, semiconductor earnings forecasts are cut or financing conditions tighten further. My subjective probability is 35%.
The third possibility is a rapid V-shaped rebound, driven by strong earnings and renewed confidence in AI spending. Such reversals happen after violent selloffs, but the scale of the current valuation and infrastructure concerns makes an immediate sustained recovery less convincing. I assign this a 20% probability.
These percentages are judgment-based scenarios rather than statistical forecasts. They are intended to frame the uncertainty, not imply precision.
One constructive signal is that weakness has been concentrated in technology. An early-November Wells Fargo assessment noted that nine of eleven S&P 500 sectors advanced in October, even as technology shares struggled. That suggests rotation rather than a uniform economic collapse, though breadth can deteriorate quickly.
The next phase depends on whether the damage spreads. If semiconductor stocks remain weak while profitable cloud and software platforms hold up, the market may be repricing AI infrastructure suppliers specifically. If selling broadens across industries and earnings estimates decline together, the risk of a prolonged bear market increases.
Investors should also distinguish profitable chipmakers with strong balance sheets from cyclical equipment suppliers and highly leveraged AI infrastructure operators. All can decline together in a panic, but their ability to survive weaker demand or expensive financing differs greatly.
The indicators worth following are hyperscaler capital-spending guidance, data-center construction delays, power availability, semiconductor order trends, bond yields, market breadth and revisions to earnings forecasts. Nvidia’s next earnings report could be an important test of confidence.
For now, I would not assume that every sharp decline is an opportunity. Waiting for several sessions of stabilization, broader participation in a rebound and evidence that earnings expectations are no longer deteriorating may sacrifice the first few percentage points of a recovery, but it can also reduce the risk of buying into a continuing slide.
The most likely outcome, in my view, remains a painful reset followed eventually by recovery. The important caveat is that the reset may not be finished. AI does not have to fail for investors in AI-related stocks to suffer substantial losses.