Microsoft plans to use fewer neoclouds as AI buildout shows signs of strain

Microsoft plans to reduce the number of specialist AI cloud providers it works with, concentrating spending on a smaller group of suppliers while prioritizing its own data centers. For builders, that signals a shift away from the broad capacity-buying strategy Microsoft used to secure GPUs during the generative AI boom. The pullback suggests Microsoft is seeking tighter control over infrastructure costs, deployment schedules, and hardware use as AI expansion strains capital budgets and power availability. A smaller supplier pool could mean larger commitments for selected neoclouds, but less predictable demand for others. Azure customers should also watch whether this consolidation affects access to GPU capacity, regional availability, or pricing as more workloads move onto Microsoft-owned infrastructure.