Microsoft guides $190B for AI capex after strong quarter; Azure +40%, AI business +123%
- Microsoft’s April 29 results were strong — revenue hit $82.9 billion, Azure grew 40%, and the company’s AI business reached a $37 billion run rate. - The number that jolted investors was spending: Microsoft now expects roughly $190 billion of 2026 capex, with fiscal Q4 capex alone above $40 billion. - That matters because demand looks real — backlog nearly doubled to $627 billion — but the AI buildout is getting vastly more expensive.
Microsoft just gave investors the cleanest version yet of the AI tradeoff. The business is working. Azure reaccelerated. The AI revenue line is getting big fast. But the bill is exploding at the same time. On April 29, Microsoft said fiscal third-quarter revenue rose to $82.9 billion, Azure grew 40%, and its AI business passed a $37 billion annual revenue run rate — then told investors to brace for roughly $190 billion in 2026 capital spending and more than $40 billion in fiscal Q4 alone. (microsoft.com) ### What was actually strong here? Pretty much the whole cloud stack. Microsoft Cloud revenue reached $54.5 billion, up 29%, while Intelligent Cloud revenue rose 30% to $34.7 billion. Azure’s 40% growth matters most because that is the business investors watch as the scoreboard for whether AI demand is turning into real cloud revenue rather than just hype(microsoft.com)% from a year earlier. (microsoft.com) ### Why did the spending number hit so hard? Because it was much bigger than the market was braced for. Microsoft spent $31.9 billion in capex in the March quarter, and management signaled that the June quarter will go above $40 billion. For calendar 2026, the company is now talking about roughly $190 billion of capital investment. That is not a normal st(microsoft.com)ower, datacenter space — not customer interest. (cnbctv18.com) ### So is demand really that strong? The backlog says yes. Microsoft’s commercial remaining performance obligation — basically contracted revenue not yet recognized — jumped 99% to $627 billion. A quarter of that is expected to convert into revenue over the next 12 months, up 39%. That is a huge pile of committed business, and it helps explain why Microsoft is willing to spend so aggressively before all the revenue shows up. (microsoft.com) ### Why isn’t Wall Street just cheering? Because investors want AI to look like software economics, not utility economics. Software is supposed to scale beautifully — sell more product, keep fat margins. But generative AI is dragging the business back toward heavy industry. You need GPUs, memory, storage, cooling, land, and power. If revenue rises 123% bu(microsoft.com). (microsoft.com) ### Is this just Microsoft, or a broader AI pattern? It looks broader, but Microsoft is the clearest case because it sits in the middle of both the cloud and the application layers. It sells Azure capacity, it sells Copilot-style software on top, and it has deep ties to OpenAI. That means Microsoft sees demand from companies building AI systems and from c(microsoft.com)t’s also why Azure growth is such a crucial signal here. (microsoft.com) ### What should investors watch next? Two things. First, whether Azure can hold around 40% growth as new capacity comes online. Microsoft said it expects Azure growth of 39% to 40% in constant currency in fiscal Q4. Second, whether margins hold up while capex keeps climbing. If demand stays ahead of supply, the spending will look justified. If growth cools before the new infrastructure earns through, the spending story gets a lot harder. (msn.com) ### Bottom line? Microsoft just showed that AI demand is real enough to justify a giant buildout. But it also showed the next phase of the AI boom may be less about clever demos and more about who can afford to pour concrete, wire datacenters, and keep buying chips at industrial scale. (microsoft.com)
