The largest technology companies in the world are spending money faster than they are making it from artificial intelligence, and the stock market has noticed. In a two-day sell-off at the end of October, Intel lost 28% of its value, Alphabet shed more than $500bn in market capitalization, Reddit cratered 29%, and Roblox dropped 18%. The Nasdaq posted its worst monthly slide since March 2025. The trigger was not a single disaster but a pattern across earnings reports. Alphabet revealed that AI-generated search summaries — the product it built to stay competitive — are cannibalizing its core search advertising revenue. The company simultaneously disclosed plans to spend up to $205bn this year on AI-related investments. Amazon reported spending more cash than it earned, with expenditures rising six-fold in three months, and committed $220bn for 2026. Intel, pivoting hard to AI chips, announced increased capital spending even as PC demand weakened and its CFO David Zinsner forecast lower-than-expected fourth-quarter revenue. The collective commitment is staggering. Meta, Alphabet, Oracle, Amazon, and Microsoft have pledged roughly $800bn in 2026 AI spending between them. That number approaches the GDP of the Netherlands. The question markets are now pricing is whether this capital will generate proportional returns or whether the industry is building cathedrals for a congregation that has not yet arrived. Reddit's 29% plunge illustrates a second-order effect that analysts are only beginning to price. The platform reported weaker daily user figures and explicitly warned that Google's AI-driven search is reducing referral traffic. This is the extraction chain in motion: tech giants build AI products that absorb traffic from the open web, but those same AI products undermine the advertising model that funds the investment. Alphabet is simultaneously the predator and the prey. Intel's position is structurally different but equally precarious. The chipmaker is increasing AI-related capital expenditure while its core PC business contracts, with Zinsner projecting the slowdown to persist through at least the first half of next year. The company does not expect to generate positive free cash flow this year. It is betting its balance sheet on a market transition it cannot yet monetize. Analysts are split on whether this constitutes a bubble or growing pains. Chris Low of FHN Financial described the situation as "pretty significant growing pains" in the AI build-out and suggested the entire sector could slow. Kathleen Brookes of XTB noted investors have "turned ruthless, punishing any signs of disappointment" and warned that if AI stocks continue to struggle, it could signal the trade has peaked. Both stopped short of calling it a bubble. Context matters: despite the sell-off, the Nasdaq remains up 7.8% over the past year, and Amazon is up nearly 8% over twelve months. Alphabet is down just 1.3% annually. The market is repricing expectations at the margin, not unwinding positions. But the structural question persists — when $800bn in committed spending meets uncertain revenue timelines, the gap between investment and return becomes the systemic risk, not any single earnings miss.
