The Twenty Minute VC (20VC) · Monday, October 26, 2026
Discussing the security risks of increasingly capable models, Satya Nadella explained that Microsoft builds zero-trust environments with agent identities, logged model calls, and auditable English chain-of-thought outputs to contain potential malfunctions.
“So that's why I kind of look at it and say, you leave AI safety outside of this... the trust architecture is going to be just, you know, what a company like Microsoft does or all the cyber [security firms].”
“It gives every agent an identity. It then logs all of its activity, all its tool calls. In fact, every model call and the COT in it. In fact, that's where I'm also asking that all the COT better be in English and not in new release.”