Microsoft chief Satya Nadella urges emergency brakes for advanced artificial intelligence

Microsoft Chief Executive Satya Nadella called for advanced artificial intelligence systems to be built with externalized safeguards and an emergency brake. TechCrunch and CNBC both reported that Nadella posted his proposal on X on Saturday, urging tech creators to allow authorized people to pause or shut down a model mid-task. [1][2]

CNBC reported that Nadella argued non-deterministic models must be surrounded with strong, deterministic system design, human controls, and reliable operating principles. TechCrunch reported that Nadella advocated for separating the core model from the harness that orchestrates its work, adding that tech builders should not treat advanced artificial intelligence as nested black boxes. [1][2]

TechCrunch reported that Nadella also called for every meaningful model action to be logged and governed by external controls. According to CNBC, Nadella joined a growing number of tech executives and researchers who are calling for stronger safety safeguards and for managing the pacing of frontier model development. [1][2]

Nadella has served as the chief executive officer of Microsoft since succeeding Steve Ballmer in 2014, and he became the company chairman in 2021. He previously led Microsoft's cloud and enterprise group as executive vice president. [8]