OpenAI Revenue Run Rate Discrepancy Triggers Selloff in Nvidia and Micron Shares

The reported figure reflects timing differences in contract recognition rather than reduced usage or cancellations. OpenAI’s own statements and enterprise contract filings show continued growth in paid seats and API volume through the third quarter. Market participants nevertheless treated the number as a demand signal, prompting immediate re-pricing across the AI hardware supply chain.

Nvidia’s data-center revenue remains tied to large-scale training clusters whose deployment schedules are set by hyperscalers and sovereign AI programs. A single model developer’s quarterly run rate therefore supplies only marginal information about aggregate silicon demand. Micron’s HBM allocation is similarly back-loaded into 2025 contracts already priced into foundry capacity.

The episode illustrates how forward-looking revenue multiples amplify small variances in private-company disclosures. Primary documents from Nvidia’s most recent 10-Q and TSMC utilization reports continue to show capacity sold out through mid-2025, independent of any single customer’s reported ARR.

Absent new primary evidence of cluster cancellations or revised capex guidance from Microsoft or Google, the price adjustment is likely to reverse once subsequent usage metrics are released.