Microsoft AI business tops $37B

Microsoft’s AI business is no longer a side story inside Azure. It is big enough to stand on its own — and big enough to reshape how people think about the whole company. The news is simple but heavy: in Microsoft’s fiscal third quarter, reported April 29, the company said its AI business topped a $37 billion annual revenue run rate, up 123% from a year earlier, while Azure and other cloud services grew 40%. (microsoft.com) ### What does “$37 billion run rate” actually mean? It is not a promise that Microsoft will book exactly $37 billion in AI revenue this year. It means the current quarter’s AI revenue, annualized, now points above that level if demand holds. That matters because it turns AI from a hype line into a business of real scale — one already comparable to large standalone software companies. (microsoft.com) ### Why is Azure the key number here? Because Azure is where a lot of the AI demand shows up first. Companies buy model access, compute, storage, and all the plumbing around AI through Microsoft’s cloud stack. Azure and other cloud services grew 40% year over year in the quarter, and Microsof(microsoft.com)rs are not only experimenting — they are renting serious infrastructure. (microsoft.com) ### Is this just OpenAI spillover? OpenAI helps, obviously, but the picture is wider now. Microsoft is selling AI through Azure, Copilot products, developer tools, security tools, and enterprise software. The company’s own framing in the quarter was that cloud and AI demand is lifting(microsoft.com) about one partnership and more about Microsoft turning AI into a full-stack sales machine. (microsoft.com) ### Are people actually using AI at that scale? Microsoft’s new Global AI Diffusion Report says yes — at least directionally. The report, published May 7, said AI usage rose from 16.3% to 17.8% of the world’s working-age population in the first quarter of 2026. That is a 1.5 percentage point (microsoft.com)ignal of momentum rather than a neutral industry census. Still, the direction lines up with what Microsoft is seeing in revenue. (blogs.microsoft.com) ### Why is Microsoft still spending so aggressively? Because AI demand is constrained by capacity. If customers want more model inference and training compute than Microsoft can supply, the bottleneck becomes chips, servers, and data centers. Microsoft has been explicit that(blogs.microsoft.com)astructure. In plain English — if the company wants to keep the growth going, it has to keep building. (microsoft.com) ### Does that create any risk? Yes. Fast AI growth looks great, but infrastructure is expensive and margins can get squeezed while capacity is ramping. Microsoft’s cloud gross margin has come under some pressure as AI infrastructure scales, even while revenue keeps climbing. So the bet here is not “(microsoft.com)d software mix to outrun the cost of all that hardware. (microsoft.com) ### What is the real takeaway? Microsoft just showed that AI is already a giant business, not a future maybe. The $37 billion run rate is the headline, but the deeper point is that Azure growth, enterprise adoption, and infrastructure spending are all moving in the same direction. That is what makes this feel durable. (microsoft.com) ### Bottom line The market has been waiting to see whether AI demand would turn into revenue at true megacap scale. Microsoft’s latest quarter says it already has. Now the question is less “is AI real?” and more “how long can Microsoft keep feeding it?” (microsoft.com)