OpenAI Revenue Reset Hits AI Stocks; Oracle Falls 5%

undefined

OpenAI's annualized revenue is about $50 billion, not $70 billion, sending Oracle down over 5% and the Nasdaq down 1.3% as AI stocks reprice growth bets.

Lead

AI stocks sold off on Thursday, October 8, after OpenAI told investors its annualized revenue was approaching $50 billion at the end of September. The figure is roughly $20 billion below the nearly $70 billion that had circulated in markets over the previous weeks. Oracle (ORCL) fell more than 5%, while Nvidia (NVDA) and AMD (AMD) declined 3-4%. The Nasdaq Composite closed about 1.3% lower.

What Did OpenAI Actually Report?

OpenAI's annualized revenue run rate is about $50 billion, up sharply from a year earlier but well short of the $70 billion estimate. The company told investors the figure stood just under that level at the end of September.

The discrepancy comes from how revenue is counted. Anthropic includes sales of its models made through cloud partners such as Amazon Web Services and Google Cloud in its annualized figure. OpenAI does not. The $70 billion estimate rested on a rough calculation that began with a figure of about $40 billion for August and added the 70% growth rate OpenAI had cited, without reconciling the two methods.

An annualized figure, or run rate, takes a recent period of revenue and multiplies it to a full year. It is a forward-looking snapshot rather than audited results, which is why definitions matter.

Why Did AI Stocks React So Sharply?

AI stocks fell because OpenAI is the largest single source of contracted demand for data-center capacity, and the lower figure narrowed the revenue base supporting that spending. Investors had treated the $70 billion run rate as evidence that customer revenue was keeping pace with infrastructure commitments.

Chipmakers followed. Nvidia and AMD supply the accelerators that fill those facilities, so any doubt over OpenAI's ability to fund purchases feeds directly into their order outlook. Both fell 3-4%, and the semiconductor weakness pulled the Nasdaq to its 1.3% loss.

Strategic Context

The gap between $50 billion and $70 billion is about accounting, not a collapse in demand. At about $50 billion, OpenAI's run rate remains many times its level of two years ago, and the company is still growing quickly.

The sell-off shows how much of the AI trade depends on a few headline metrics. Revenue run rates are self-reported, defined differently across companies, and rarely audited. A single revision of $20 billion, about 29% of the earlier figure, was enough to move trillions of dollars in market value on a single session.

The episode also underscores the mismatch between spending and revenue. OpenAI and its partners have committed to data-center and chip purchases measured in hundreds of billions of dollars over several years. That spending assumes revenue will grow into the commitments, so a lower starting point raises the required growth rate.

What Comes Next for AI Infrastructure Spending?

Spending plans are unlikely to change immediately, but scrutiny of the revenue behind them will increase. Long-term compute contracts, construction timelines and chip orders were set against growth targets that remain in place, and the revised run rate does not by itself cancel any of them.

The nearer-term focus is on disclosure. Investors in Oracle, Microsoft and the chip suppliers will look for how much of their backlog depends on OpenAI and on what payment terms. Earnings calls in the coming weeks give those companies a chance to address the exposure directly. Standardized reporting of annualized revenue, including treatment of cloud-partner sales, is likely to draw more attention as AI developers approach public markets.

Outlook

OpenAI's revenue is growing quickly, but at roughly $50 billion annualized it sits $20 billion below the figure markets had priced in. The reset hit Oracle hardest, pulled Nvidia and AMD lower and dragged the Nasdaq down about 1.3%. The next test for AI stocks is whether upcoming company disclosures show demand and funding sufficient to support the infrastructure build-out already under way.