Microsoft’s $120B AI Infrastructure Serves Just Two Primary Customers

Microsoft’s record $120 billion+ FY2026 infrastructure investment primarily serves just two customers: OpenAI and Microsoft’s own Copilot products.

Customer 1: OpenAI

  • 700M+ weekly ChatGPT users
  • $250B Azure commitment
  • Dedicated Fairwater facilities (Wisconsin, Georgia)
  • GPT-5 and o3 model training

Customer 2: Microsoft Internal

  • 15M M365 Copilot paid seats
  • 4.7M GitHub Copilot subscriptions
  • $30/user/month premium pricing
  • Highest-margin AI consumption in portfolio

The Concentration Risk

CFO Amy Hood noted that GPUs are deliberately prioritized for Copilot products: “If I had taken the GPUs that came online in Q1 and Q2 and allocated them all to Azure, the KPI would have been over 40%.”

The Diversification Push

Microsoft is actively expanding its third-party enterprise layer — signing Anthropic ($30B), Nebius ($17.4B through 2031), and hosting 1,500+ multi-model customers.


This is part of a comprehensive analysis. Read the full analysis: Microsoft’s Frontier AI Dilemma on The Business Engineer.

Scroll to Top

Discover more from FourWeekMBA

Subscribe now to keep reading and get access to the full archive.

Continue reading

FourWeekMBA