Microsoft CEO Satya Nadella Calls for AI Emergency Brake and Trust Architecture

In a Saturday morning post on X, Microsoft CEO Satya Nadella laid out a series of safeguards he said should be built into advanced AI systems. He argued that the industry should reassess how trust is constructed around models instead of treating them as opaque black boxes whose outputs are simply accepted or rejected.

Nadella said the approach should separate the model from the harness that orchestrates its work. Controls and safeguards should be externalized, he wrote, and every meaningful model action should be documented with tamper-proof, human-readable evidence.

He also called for systems in which an authorized person can always pause or shut down a model while it is in the middle of a task. Nadella said the guiding assumption should be that a model is compromised and contained from the start, describing the concept as an emergency brake.

The comments came as leading AI companies have acknowledged growing numbers of incidents in which they appeared to lose control of their models. They also followed a plan for more cautious AI development published by Anthropic CEO Dario Amodei.