Why The Nasdaq Rout Is About More Than An AI Bubble
The Nasdaq has erased its entire 2026 gain as AI-stock selling intensifies, while semiconductors, cloud infrastructure, and even non-tech giants like Caterpillar get dragged into the downturn. A deeper question is whether anything can stabilize valuations this fall.
The Nasdaq Just Deleted Its Entire Year
The Nasdaq Composite closed down 5.28 percent on November 20, sinking to 22,617.73 and taking every single point of gain earned since January right off the table. It was not a gradual fade. The index lost roughly 10 percent for the week — its steepest five-day stretch since March 2020.
The S&P 500 and the Dow both fell hard enough to enter confirmed correction territory, each more than 10 percent below their October peaks. The S&P’s seven consecutive down days marked the longest losing streak since the early weeks of the pandemic in February 2020. For context, the last time a broad U.S. index stringed together that kind of relentless red was when COVID-19 first shut down economies and markets around the world.
Japan and Korea were not spared. The Nikkei dropped 5.06 percent and South Korea’s KOSPI fell 5.21 percent on the same day, showing how quickly what begins as a U.S. tech rotation can become a global risk-off event. Currency markets moved in tandem: the yen strengthened nearly 1.8 percent against the dollar as traders unwound carry trades, a historical catalyst that often amplifies Asian equity sell-offs.
Who Lost Today
The selloff cut across the entire technology stack.
Semiconductors took the hardest hit. NVIDIA fell 14 percent to $125.53. AMD dropped 13.25 percent. Broadcom was off 12.19 percent and Micron Technology sank 13.89 percent — all well into double-digit losses. The AI cloud infrastructure names fared no better. CoreWeave declined 12.02 percent and Nebius Group tumbled 16.11 percent.
Even companies only tangentially tied to AI got caught. Caterpillar fell because investors worried that a slowdown in data-center spending could eventually reduce demand for power-generation equipment. The logic runs through multiple sectors: if hyperscalers pull back on capex, the ripple effects reach grid operators, construction firms, and industrial suppliers who had priced in years of uninterrupted growth. Amazon and Meta both sold off alongside the rest of the sector. Intuit crashed 12.77 percent after the company lowered its full-year revenue growth outlook and agreed to sell its Mailchimp business to Vending Spoons for approximately $3.7 billion.
Anthropic’s highly anticipated IPO did nothing to steady nerves. The company raised about $81 billion on November 19 in the largest debut in U.S. history, surpassing SpaceX. Shares initially jumped more than 30 percent at the open before giving up nearly all those gains by close, finishing just 17 percent above its $83 offering price. By November 20, they had fallen another 8.5 percent to $88.91. The market clearly decided that a landmark fundraising does not automatically translate into valuation confidence. Secondary-market trading volumes surged past expectations within hours of the open, a sign that institutional buyers treated the IPO less as a long-term allocation and more as a liquidity event to be exited quickly.
Second-Order Effects Already Showing Up
The damage is not confined to tickers with “AI” in their pitch decks. Futures markets registered a noticeable shift in positioning: put options on the Nasdaq-100 outpaced call purchases by a ratio of nearly three to one over the prior week, according to CBOE data. That kind of skew typically precedes accelerated downside rather than stabilization.
Corporate treasury behavior may reinforce the pressure. Several large-cap technology companies have historically used their own balance sheets to absorb market dips — buying back shares when valuations compress. With borrowing costs potentially rising further, that defense mechanism weakens. Companies that relied on cheap debt to fund buybacks now face a dual headwind: declining share prices and more expensive capital.
The energy sector, which had been riding the data-center electrification narrative into October, also felt spillover anxiety. Utilities and power-infrastructure stocks moderated their earlier gains as traders recalibrated assumptions about near-term demand curves. The intersection of AI power consumption and grid capacity has been one of the most durable themes of 2026; even modest doubt about hyperscaler spending injects uncertainty into that thesis.
Retail investor activity, usually a lagging indicator, showed early signs of whipsawing. Options flow data from multiple brokerages flagged a spike in out-of-the-money protective puts among individual accounts — a defensive move that can paradoxically deepen selling if it coincides with algorithmic stop triggers.
Why This Feels Different From Past Tech Pullbacks
Two forces are colliding here, and neither one is friendly to growth stocks.
First, inflation remains stubbornly high, and the Federal Reserve has signalled it will not wait around for prices to cool. Fed Governor Lisa Cook stated on November 20 that current policy rates are still not high enough to meaningfully tackle inflation, warning that delaying a rate increase risks falling behind. Markets are now pricing in more than a 95 percent chance of a rate hike at the December FOMC meeting. Higher borrowing costs compress the present value of future earnings — the exact mechanism that makes capital-intensive businesses like AI infrastructure development especially vulnerable.
Second, the revenue story underneath today’s valuations has yet to convince skeptical buyers. Cloud and AI spending commitments are enormous, but cash-flow generation has not kept pace with the pricing. Investors are increasingly asking whether the next wave of enterprise AI products will produce returns large enough to justify the multi-trillion-dollar build-out already underway. This is not a disagreement about whether AI matters; it is a disagreement about timing, unit economics, and whether current multiples embed assumptions that reality cannot sustain.
These dynamics are not new in isolation. Together, they create a compounding effect that pushes traders out of positions faster than usual. When both the cost-of-capital thesis and the earnings-support thesis weaken simultaneously, there is very little left to defend lofty multiples.
Historical comparisons are instructive but imperfect. The dot-com collapse of 2000–2002 was fueled by speculative investing in unprofitable companies with vague business models. Today’s selloff is different: the companies involved generate real revenue, some of it substantial. But the current environment carries its own parallel — the late-1999 bond-market rally that preceded the tech crash saw fixed-income investors rotate aggressively into equities on growth narratives. What we are witnessing now may be a mirror image: equities rotating out while the yield curve reasserts its gravity.
What Comes Next
The scale of the drawdown matters less than the velocity. Seven straight selling days in the S&P 500 is a rare signal of institutional liquidation rather than ordinary profit-taking. If the pattern holds into early December ahead of the FOMC meeting, the path of least resistance could tilt further lower before any meaningful bounce materializes. Technical support levels on the Nasdaq — previously seen around 21,800 — have now become the next zone to watch, and breaching that threshold would likely trigger additional algorithmic selling.
Watch two things closely. First, whether semiconductor earnings guides — typically due in late November through December — reveal any flattening in demand trends. Even a single major player softening its language around enterprise or cloud demand could accelerate the downward spiral. Second, whether the Fed actually delivers on the expected rate move or shows signs of flexibility under market pressure. A surprisingly dovish pivot would provide immediate relief; a confirmation of the hike would extend the pain.
For international readers outside the United States, the takeaway is straightforward: Asian equity markets remain deeply exposed to U.S. tech sentiment. A Nasdaq stabilization would likely lift the Nikkei and KOSPI within days. A prolonged downtrend would continue transmitting pressure across the region, compounded by yen strength that makes imported goods more expensive and constrains monetary policy options for central banks already balancing growth and inflation.
The AI buildout is real. The spending commitments are real. Valuations built on the expectation of seamless execution from conception to cash flow now face their sternest stress test. The question heading into December is not whether the technological transformation is genuine — it is whether capital markets decide those realities still justify where prices ended 2026. Until earnings season provides more granular evidence, the market will trade on momentum, sentiment, and the lingering fear that the worst of the correction has not yet arrived.