tech
Executive Briefing: Your AI vendor contract isn't built for a capacity crunch. 3 prompts to fix it before your budget meeting
Microsoft plans to spend roughly $190 billion this year and still expects to run short on capacity. And Microsoft is not alone: across the four biggest hyperscalers, combined 2026 capital spending is on track to approach $700 billion, nearly double what they spent in 2025.
TL;DR
- Microsoft's capital expenditure for AI infrastructure is projected to be around $190 billion this year, with significant ongoing investment expected from major hyperscalers.
- AI's demand for physical resources (chips, memory, power, data centers) transforms it into an industrial business, unlike traditional software models.
- AI vendor agreements are evolving into supply contracts due to hyperscalers' scale of spending and ongoing capacity rationing.
- Key considerations for AI vendor contracts now include allocation, capacity terms, and fallbacks.
- Forecasting AI demand should shift from 'seats' to 'tokens' to accurately budget for computational resources.
- Companies need to audit their AI supply chains and implement new forecasting and routing strategies to manage costs and capacity effectively.