Microsoft CEO Backs Open-Weight AI, Urges Multi-Model Strategy

Enterprise AI is increasingly built around proprietary models sold by a small group of providers, creating a risk that customers surrender not only fees but also the prompts, feedback and workflow context that sharpen those systems. Microsoft CEO Satya Nadella, whose company is a major OpenAI partner, argues that businesses should separate their data and orchestration layers from any one model. The issue matters because control over interaction records and metadata can determine whether companies retain intellectual property, bargaining power and the ability to switch suppliers.

In a July 13, 2026 warning, Nadella said companies can effectively pay twice for AI — once in cash and again through proprietary knowledge — when vendors reserve rights to learn from usage and interaction data. He urged enterprises to retain those records and use independent orchestration layers or AI gateways to route work across proprietary and open-weight models. On July 24, Microsoft joined NVIDIA, Meta, IBM and 21 other technology companies in an open letter urging U.S. policymakers to avoid premature restrictions on open-weight AI.