Microsoft cuts 2026 spending forecast as outside AI capacity arrives late

Microsoft lowered its 2026 capital-expenditure forecast to $170 billion from $175 billion, a $5 billion reduction. Capital expenditure funds long-lived assets such as data centers and servers; CFO Amy Hood attributed the cut to delays in third-party computing capacity.^[1]

A separate report, citing Hood's comments on the earnings call, said delays at third-party AI-chip providers would push some available capacity into the first half of 2027. That capacity is the computing power needed to run AI models.^[2]

Delivery dates determine when these investments can start producing revenue. With some capacity pushed into the first half of 2027, Microsoft must wait longer to use those resources to expand AI services, while suppliers whose revenue depends on delivery may also wait longer to earn it. That puts pressure on the build-out chain: money is committed before equipment can begin paying for itself.^[1][2]

For specialized AI-cloud providers such as CoreWeave, Nebius and IREN, financing and operating costs during that wait can squeeze margins. Reports said investors were already concerned that rising interest rates would make their expansion more expensive, and their shares fell after Microsoft's spending came in below expectations.^[3]The 10-year US Treasury yield stood at 4.98% on October 29, down 7 basis points, or 0.07 percentage points, from 5.05% the previous session; that decline in a borrowing-cost benchmark does not remove the revenue delay caused by late delivery.

Impact

What to watch next

• First half of 2027: watch whether the delayed capacity becomes available within Hood's stated window. Delivery would ease the risk of further revenue delays for suppliers; another slip would extend the cost burden.^[2]