Microsoft hits $37B AI run rate

Microsoft’s latest quarter matters because it turns an AI story into a money story. For two years, the big question around generative AI has been simple — are these companies building expensive infrastructure faster than customers are willing to pay for it? On April 29, Microsoft gave the clearest yes yet. It said its AI business is now running at a $37 billion annual revenue pace, up 123% from a year earlier, while Azure kept growing fast enough to show the demand is not hypothetical. (microsoft.com) ### What exactly did Microsoft report? Microsoft’s fiscal third quarter, which ended March 31, 2026, came in strong almost everywhere that matters. Revenue rose to $82.9 billion, operating income hit $38.4 billion, and diluted EPS reached $4.27. Intelligent Cloud revenue climbed to $34.7 billion, and Microsoft Cloud(microsoft.com)dard segment, but big enough now that management highlighted it on its own. (microsoft.com) ### What does “$37 billion run rate” mean? It does not mean Microsoft booked $37 billion of AI revenue in one quarter. It means the company is annualizing its current pace of AI revenue — basically taking the latest quarter’s level and expressing it as a yearly rate. That matters because it gives investors a cleaner(microsoft.com)as moved beyond pilot projects and into production spending across cloud infrastructure and software. (microsoft.com) ### Why is that such a big deal? Because the market has been treating AI capex like a test. Microsoft, Meta, Alphabet, and others have been pouring tens of billions into GPUs, data centers, networking, and power. The fear was that demand might be real but not monetizable fast enough. Microsoft’s number does not answ(microsoft.com) demand into revenue at serious scale. (microsoft.com) ### Where is the demand showing up? A lot of it runs through Azure. Microsoft guided Azure growth of 39% to 40% in constant currency for the next quarter, ahead of what analysts were expecting. It also said Microsoft 365 Copilot now has more than 20 million paid commercial seats. Put those together and you get the s(microsoft.com)s important because software margins are usually better, but cloud infrastructure is what locks customers in first. (cnbc.com) ### So why did investors still focus on spending? Because the bill is enormous. Microsoft said fiscal 2026 capital expenditures will reach about $190 billion, far above prior expectations, and CNBC noted management tied a big chunk of that increase to higher component costs, especially memory. Third-quarter capex and finance le(cnbc.com)sappearing — from “is AI demand real?” to “how durable are returns if the build-out stays this expensive?” (cnbc.com) ### Is this mostly an OpenAI story? OpenAI is part of it, but not the whole thing. Microsoft’s filings and earnings materials still show how intertwined the companies are, yet the more important point now is distribution. Microsoft has Azure, GitHub, Microsoft 365, Dynamics, and a giant enterprise sales channel. That means it c(cnbc.com)oftware upgrades. The OpenAI tie-up helped start the wave, but Microsoft is trying to turn that advantage into a broad platform business. (microsoft.com) ### What’s the real takeaway? Microsoft did not just say AI is promising. It put a very large revenue number on the table. That does not make the spending risk vanish — $190 billion in capex is still staggering — but it changes the tone of the conversation. The question is no longer whether Microsoft can sell AI at scale. Now it is whether that scale can stay ahead of the cost of building the machine. (microsoft.com)