Technology stocks closed out October with their worst monthly slide since March 2025, as a two-day rout driven by runaway AI spending fears erased hundreds of billions of dollars in market value and left even the strongest AI promoters defending their balance sheets.
According to a BBC News report from October 30, shares in Intel plummeted 28% on the Friday session — shedding more than $130 billion in value — while Alphabet fell another 1.4% after a 12% plunge the day before. Amazon and Reddit also declined, and the Nasdaq index of the largest US-listed technology firms ended the month deeply in the red. The trigger was not a single disaster but a cumulative realization: the industry's AI bills are coming due faster than its AI revenues. We are following the selloff and its fallout in our AI business coverage.
The Numbers Behind the Rout
The BBC reported that Meta, Alphabet, Oracle, Amazon and Microsoft have collectively committed to spending $800 billion in 2026, with the bulk directed at AI infrastructure. Earnings season turned those commitments from an abstract ambition into a line item investors could no longer ignore — because at several companies, the earnings have not kept pace.
Alphabet's Thursday plunge wiped more than $500 billion off its market value after the company revealed that the AI summaries now placed at the top of search results are eating into its core search advertising revenue. The irony was not lost on observers: the company at the center of the AI boom took one of its biggest hits because its own AI is disrupting its main business. Alphabet still expects to spend up to $205 billion this year on investments, almost entirely AI-related.
Intel's collapse was more straightforward. The chipmaker said it planned to increase AI spending despite weaker demand for the computers that buy its traditional processors. Chief financial officer David Zinsner told investors the firm did not expect to bring in more cash than it spent this year and forecast lower-than-expected fourth-quarter revenue. "Looking ahead, we see the PC slowdown continuing at least through the first half of next year," he said in the company's third-quarter earnings report.
Amazon, meanwhile, disclosed it had also spent more cash than it was making, with spending up six-fold over the span of three months. The company has committed $220 billion this year, much of it toward AI-adjacent infrastructure.
The Bubble Question Returns to the Center
The selloff has revived the question that has hovered over the AI boom all year: whether hundreds of billions in data center construction can ever be repaid by AI revenues, or whether the industry is building ahead of demand the way telecom built fiber ahead of traffic in 1999.
The market stress is now showing up in private markets too. The Wall Street Journal reported that OpenAI raised less than half of the $30 billion target for its latest funding round — a striking data point for the most valuable private AI company, and one that follows the Financial Times' report that OpenAI had slashed the round's target by $10 billion as anxieties about AI spending mounted. When the sector's flagship fundraiser is marked down mid-round, public investors notice.
There were scattered bright spots in the gloom. With Nvidia and other AI infrastructure names sold off, several outlets including TechPowerUp and TweakTown reported that Apple briefly reclaimed its position as the world's most valuable company — largely because its strategy of avoiding massive AI capital expenditures suddenly looked like discipline rather than timidity.
Reuters described Wall Street closing sharply lower as AI "cash burn" fears fueled the tech selloff, and 24/7 Wall St. noted that the Nasdaq 100 briefly fell into correction territory during intraday trading — milestones that mark a shift in tone for a market that spent most of the year rewarding AI spending without scrutiny.
What Happens Next
October's correction does not settle the debate over whether AI spending is prudent investment or overbuild. Hyperscalers continue to argue that compute capacity is the limiting factor on their AI businesses, and several reaffirmed their spending plans in the same breath as their cost disclosures. The bond markets have started to weigh in as well, with the financing costs of AI data center debt creeping higher.
What changed this week is simpler: investors stopped treating AI capex as a free good. Companies that show revenue following their spending will be forgiven; companies that show spending running ahead of revenue, as Intel and Alphabet did from different directions, are being repriced in real time.
For the AI industry, November now carries a clear test. The spending commitments are locked in for 2026 — the $800 billion and change is largely earmarked. Whether the revenues arrive on schedule will determine if October's selloff was a healthy correction or the first chapter of something longer.
---
Stay Ahead of AIGet the latest AI news, analysis, and breakthroughs — all in one place.
Read more AI news →
