Post

Nvidia’s launch materials indicate that it has AI safety and security collaborations with dozens of other tech companies. The company says Anthropic and Nvidia are “building security into Claude Managed Agents.” Only one notable name is missing entirely from Nvidia’s list: OpenAI.

wired.comNvidia’s Answer to Rogue Agents Is an Open-Source AI Security SystemIn the wake of a series of high-profile AI safety incidents, Nvidia is introducing a new software tool that helps keep agents from escaping containment.

The chip giant has also developed a new software platform called Sentry, an isolated security domain for chips that’s supposed to continuously monitor long-running AI agents. Sentry can act in addition to OpenShell to “quarantine agents that attempt to move outside their boundaries.”

wired.comNvidia’s Answer to Rogue Agents Is an Open-Source AI Security SystemIn the wake of a series of high-profile AI safety incidents, Nvidia is introducing a new software tool that helps keep agents from escaping containment.

Why didn’t George W Bush ask al Qaeda for advice to handle international terrorism? We should ask the drug cartels on how best to handle the international drug trade! This CEO punk and his accomplices need to be arrested

Funny how I watch a 4k dvd and it talks about how piracy isn’t a victimless crime, and yet these AI mobsters committed the largest act of piracy in human history and we get articles peddling their advice

Open-sourcing the enforcement layer is the right call — security through obscurity never worked for sandboxes. The real test: will frontier labs actually adopt it, or ship their own incompatible versions?