Friday, Nov 13 | --:--
Back to home

Microsoft’s AI Run Rate Hits $37B — Cloud RPO Nearly Doubles

Microsoft FY26 Q3: $82.9B revenue (+18%), AI annual run rate $37B (+123%), Microsoft Cloud $54.5B (+29%), commercial RPO $627B (+99%). The trade is AI monetization at scale with Azure +40% — vs still-opaque AI margin mix and OpenAI-related non-GAAP noise.

Times of AI Desk 5 min read Redmond View as Markdown
Microsoft earnings chart showing AI revenue growth and cloud performance for FY26 Q3 announced April 29, 2026

The useful number is not total revenue. It is the $37 billion AI annual run rate (+123% YoY) sitting inside a cloud machine whose commercial remaining performance obligation nearly doubled — a rare public scale check on whether frontier AI spend is turning into booked demand.

Microsoft (April 29) reported FY26 Q3 (ended March 31, 2026): total revenue $82.9 billion (+18%, +15% constant currency), operating income $38.4 billion (+20%), net income $31.8 billion (+23% GAAP), diluted EPS $4.27. CEO Satya Nadella: AI business “surpassed an annual revenue run rate of $37 billion, up 123% year-over-year.”

What shipped (earnings primary)#

LinkEmbed

Metric Figure (Microsoft IR)
Total revenue $82.9B (+18% / +15% CC)
Microsoft Cloud $54.5B (+29% / +25% CC)
AI ARR $37B (+123% YoY)
Commercial RPO $627B (+99%)
Azure + other cloud +40% (+39% CC)
Shareholder return $10.2B dividends + buybacks

Segments: Productivity and Business Processes $35.0B (+17%); Intelligent Cloud $34.7B (+30%); More Personal Computing $13.2B (−1%). Inside Productivity: M365 Commercial cloud +19%, Consumer cloud +33%, LinkedIn +12%, Dynamics 365 +22%.

Non-GAAP adjustments primarily reflect impacts from OpenAI investments — according to the company’s disclosure.

Claims vs checks#

Figures are Microsoft Investor Relations primary. The $37B AI run rate is a company-defined aggregate (infrastructure + applications); Microsoft does not publish a fully audited AI P&L or margin stack in the release. Treat ARR growth as a vendor commercialization signal, not independent unit-economics proof. Azure +40% is the infrastructure demand read; RPO +99% is the backlog read.

Limits#

  • AI run-rate methodology (what counts as “AI revenue”) is Microsoft-defined.
  • OpenAI investment effects blur non-GAAP comparisons.
  • MPC decline shows AI lift is uneven across the portfolio.

Sources#

Scroll to continue reading