Satya Nadella says we should assume all AI models are ‘compromised’

In a lengthy post on X, Microsoft CEO Satya Nadella warned about the risks posed by highly advanced AI models and outlined steps to address them. He argued that treating AI as a set of opaque black boxes is no longer acceptable and called for more transparent systems that can contain, monitor, and produce tamper‑proof, human‑readable evidence. His suggestions echo other calls from the AI safety community.

Summary written by Kernelia from the original article by The Verge — AI. The story and its rights belong to its author.