The Verge AI · 10/11/2026, 06:10:17
Microsoft CEO Nadella Proposes Default Compromise Model for AI Safety and Containment
Microsoft CEO Satya Nadella advocates moving beyond treating AI as 'nested black boxes' toward transparent systems with tamper-proof evidence. He proposes a radical security assumption: all advanced AI models must be treated as compromised from the start, requiring immediate containment and monitoring akin to an emergency brake. This shift aims to enforce zero-trust principles against superintelligence risks.
SOURCE COVERAGEFull Original Report
Part OfThe AI Superintelligence Slowdownsee all updatesTerrence O'Brien is the Verge’s weekend editor. He’s covered the tech industry for over 18 years and knows a thing or two about synths.
In a lengthy post on X, Microsoft’s CEO laid out his views on the dangers posed by highly advanced AI models and how to confront those risks. Nadella says we can no longer accept a world where AI is treated as a “set of nested black boxes” whose advice and actions we simply accept or reject. He calls for building a more transparent system where models can be contained, observed, and leaves behind “tamper-proof human readable evidence.”
Many of his recommendations align with what we’ve heard from others in the industry: timely incident disclosure, independent audits, verifiable data, and containment. It’s on that last point that he appears to go slightly farther than some others in the field, saying:
We must assume a model is compromised and contain it from the start. Think of it like an emergency brake. An authorized person should always be able to pause or shut down a model mid-task. More advanced models will require more advanced containment technologies that we need to standardize on.
Unfortunately, Nadella also refers to AI as “super intelligence” throughout the post.
Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates.
- Terrence O'Brien
More in: The AI Superintelligence Slowdown
Anthropic is cutting off its internal evaluations from the internetTerrence O'Brien2 PM UTCAnthropic published a report about investigating “unintended model actions” during “evaluations and internal use.”Jay Peters12 AM UTCAnthropic’s AI gave Philadelphia police a fake tip about an unsolved homicideEmma RothOct 9