AI Revenue Estimates Differ From Cloud and Data Center Sales

Private AI companies and public infrastructure providers report figures that describe different kinds of sales, making the totals difficult to compare directly.

OpenAI’s annualized run rate was $40 billion in August 2026 and neared $50 billion at the end of September. A separate comparison approaching $70 billion used a grossed-up calculation that included partner sales to align the accounting with Anthropic’s. That figure uses a different basis and is not a like-for-like revision to the September estimate.

Anthropic’s run rate crossed $47 billion earlier in May 2026 and exceeded $65 billion by the end of July. Its third-quarter 2026 revenue was $16.8 billion. The July and quarterly figures appeared in investor materials rather than public company filings. An annualized run rate extends a recent sales level across 12 months; it does not represent revenue booked during a completed year.

Alphabet’s Google Cloud recorded $24.8 billion in Q2 2026 revenue, 82% higher than a year earlier. The quarter closed June 30. Its segment combines AI infrastructure and solutions with other cloud services.

Amazon Web Services (AWS) generated $42.2 billion in Q2 2026 revenue, up 37% from a year earlier for the quarter that closed June 30. Amazon separately put AWS’ AI business above a $25 billion annual revenue run rate. That annualized measure is distinct from AWS’ quarterly revenue.

Microsoft Cloud revenue reached $59.3 billion in Microsoft FY26 Q4, which ended June 30, a 27% year-over-year increase. The measure includes Microsoft 365 Commercial cloud, Azure and other cloud services, commercial LinkedIn, and Dynamics 365. Nvidia recorded $89 billion in data-center revenue for fiscal Q2 2027, which ended July 26, up 117% from a year earlier. Nvidia does not report that segment as AI-only revenue.

The public-company totals combine non-AI operations with AI-related services or infrastructure and reflect completed quarters. The private labs’ figures instead project a recent sales pace across a year. Those differences in business scope, accounting basis and time period mean the numbers offer context on the sector’s scale, not a direct ranking of AI revenue.