Microsoft says its AI business has passed a $37 billion run rate

Microsoft said on April 29 that its AI business has passed a $37 billion annual revenue run rate.

The important signal in Microsoft’s Q3 results is not just growth. It is that AI is now being reported as a large, measurable business with 123% year-over-year growth. Microsoft Cloud revenue reached $54.5 billion for the quarter, up 29%, while Azure and other cloud services grew 40%. Commercial remaining performance obligation rose 99% to $627 billion.

For Norwegian executives, this means AI can no longer sit only in an innovation budget. Microsoft is using AI to pull more of the enterprise workspace into Azure, Microsoft 365, Dynamics and the Copilot layer. That creates real productivity upside, but also deeper vendor dependency, more complex licensing and stricter requirements for data classification.

The fact is that Microsoft is growing across Productivity and Business Processes, Intelligent Cloud and Azure. The judgement is that CIOs should treat the numbers as a negotiation signal: Copilot and agent features will be bundled more tightly into core enterprise agreements. Architecture choices, exit options and cost controls matter before the next renewal, not after it.

Boards and CFOs should ask for three things now: a map of which Microsoft AI features are actually enabled, a cost model for scaling from pilots to broad use, and a data-risk assessment for Copilot, agents and integrations with business systems. The productivity gain does not come automatically because the license exists. It comes when processes, access control and measurement are in place.

The practical advice is simple: use Microsoft’s own numbers as a reminder of market speed, but do not let a vendor roadmap become the company’s AI strategy. Keep optionality in the model layer, data platform and agent architecture where the risk is high.