Microsoft CEO Satya Nadella Calls for AI Emergency Brakes and a New Trust Architecture

Microsoft CEO Satya Nadella called for a new trust architecture in artificial intelligence, including an emergency shutdown mechanism for advanced systems. Writing on X, Nadella argued that society cannot treat powerful artificial intelligence systems as closed black boxes whose decisions are simply accepted or rejected. He proposed separating the underlying model from the harness that orchestrates its work, externalizing safeguards, and documenting every meaningful model action with tamper-proof, human-readable evidence.

Furthermore, he stated that authorized personnel must always retain the ability to pause or shut down a model mid-task while assuming that systems are compromised from the start. Nadella's proposal matters as leading artificial intelligence companies increasingly acknowledge incidents where they appeared to lose control of their models.

His comments follow a similar cautious development plan published by Anthropic CEO Dario Amodei, highlighting a growing industry focus on tighter containment and oversight.