Microsoft plans $190B capex

- Microsoft told investors on April 29 it expects roughly $190 billion of 2026 capital spending as it races to expand AI datacenter capacity. - The key detail was Amy Hood’s breakdown: about $25 billion of that 2026 total comes from higher component pricing, not just more servers. - That matters because Microsoft’s AI buildout is now big enough that chip and memory inflation can reshape capex plans.

Microsoft’s news here is not just “AI spending stays high.” It’s that the bill got a lot bigger — and Microsoft said out loud why. On its April 29 fiscal third-quarter earnings call, the company told investors to expect roughly $190 billion in calendar 2026 capital expenditures. That is an enormous number even by hyperscaler standards, and Microsoft tied a big chunk of it to higher component prices, not only to demand growth. (microsoft.com) ### What did Microsoft actually say? Amy Hood, Microsoft’s finance chief, said the company expects to invest roughly $190 billion in capital expenditures in calendar 2026. She also said the mix of short-lived assets should stay similar to the March-quarter pattern. In the quarter that ended March 31, Microsoft reported $31.9 billion of capital expenditures and finance leases, up 49% year over year. (microsoft.com) ### Why is $190 billion such a shock? Because Wall Street was nowhere near that level. CNBC noted the forecast came in well above consensus, with analysts looking for something closer to the mid-$150 billions. So this was not a routine nudge higher — it was Microsoft telling the market that the AI infrastructure race is still accelerating and that the cost base is moving with it. (cnbc.com) ### Is this all just more datacenters? Not exactly. More capacity is the main story, but Microsoft also said pricing is doing real damage. Hood said the 2026 capex figure includes about $25 billion from higher component pricing. That means a meaningful slice of the increase is inflation inside the AI supply chain — memory, networking gear, accelerators, and the other expensive pieces that go into AI clusters. (microsoft.com) ### Why do component prices matter so much? Because AI infrastructure is unusually concentrated in a few brutal line items. If memory pricing jumps, or high-end server parts stay tight, the total bill moves fast. This is different from a normal software story where growth mostly hits operating expense. Here, Microsoft is say(microsoft.com)hink about margins, free cash flow, and depreciation later on. That last part is an inference from the capex breakdown and Microsoft’s margin guidance. (microsoft.com) ### But is the business strong enough to justify it? For now, yes. Microsoft posted $82.9 billion in quarterly revenue, up 18%, with Microsoft Cloud revenue above $54 billion and Azure growth beating expectations. Satya Nadella also said Microsoft’s AI business surpassed $37 billion in annual recurring revenue. So the company is not spending into a slowdown — it is spending into demand that still looks very real. (news.microsoft.com) ### Why did the stock reaction feel mixed? Because the quarter itself was strong, but the forward picture had some softness. Microsoft’s revenue guidance for the next quarter landed a bit below consensus, and implied operating margin also looked lighter than some investors expected. So the market had to weigh two things at once — stronger AI demand and a much bigger infrastructure bill. (cnbc.com) ### Is this just a Microsoft story? No — it’s a read-through for the whole AI stack. If Microsoft is signaling that component inflation alone adds about $25 billion to next year’s capex, that suggests the supply chain is still tight even at hyperscaler scale. Basically, the AI boom is no longer just about who wants capacity. It’s also about who can afford the inputs. (microsoft.com) ### Bottom line Microsoft didn’t just raise the AI spending bar. It showed that the bar itself is getting more expensive to build, and that may be the more important signal.

Get your own daily briefing

Scout delivers personalized news, insights, and conversations tailored to your role and industry.

Download on the App Store

Shared from Scout - Be the smartest in the room.