Yes, OpenAI is a big chunk — but even excluding it, the commitments still grew 28%.

Cloud momentum in addition to the mega-deals

CFO Amy Hood said ~45% of total RPO is driven by OpenAI. The remaining ~$345B (+28%) reflects what she called “ongoing broad customer demand across the portfolio.”

Azure itself grew impressive 39% YoY, beating guidance.

Microsoft Cloud revenue crossed $51.5B this quarter. Using my previous conservative 45% Azure share proxy, that implies ~$90B+ Azure revenue annualized.

But Azure growth could have been even higher. With demand still exceeding capacity, Microsoft is rationing compute across M365 Copilot, GitHub Copilot, R&D, Teams, and Azure.

Satya Nadella was clear on the strategy: “We don’t want to maximize just one business... [we want to] build the best LTV portfolio.”

Amy Hood noted that if the GPUs that came online were allocated entirely to Azure, growth would have been “over 40.”

The bigger strategic play: agents as the next platform

Satya suggested to “[think of] agents as the new apps” and laid out the full stack requirements: model catalog, tuning, orchestration, context engineering, safety, observability, and security.

Side note (maybe I’m biased): that stack description reads a lot like what marketplaces do at scale — catalog + distribution + governance + trust — just applied to agents.

Agent adoption is already meaningful:

• Over 80% of Fortune 500 companies have active agents built using Microsoft’s low-code tools.

• M365 Copilot hit 15 million paid seats, with seat adds up 160% YoY.

• Larger deployments are accelerating — the number of customers with 35,000+ seats tripled, including one deployment at 95,000 seats.

The control-plane move is particularly interesting.

Microsoft introduced Agent 365 to extend governance, identity, and security controls to agents — even those running on other clouds.

Adobe, Databricks, SAP, ServiceNow, and Workday are already integrating.

As Satya framed it: “We are the first provider to offer this type of agent control plane across clouds.”

Governance is scaling alongside AI adoption

One datapoint shows the governance layer is growing with usage: 24B Copilot interactions audited by Purview (+9x YoY).

This signals where enterprise AI spending will increasingly concentrate: not just model access, but compliance, auditability, and risk management tooling.

AI infrastructure buildout is remarkable

Microsoft spent $37.5B on CapEx this quarter. That’s ~$150B annualized pace if continued — though management expects CapEx to step down sequentially next quarter.

Still, Microsoft is compounding two engines at once: strong cloud growth and a rapidly expanding infrastructure layer.

This accelerates its ecosystem — more production workloads, more partner attach, more marketplace motion.

Are you leveraging it?