Nvidia unveils security platform to stop AI agents from going rogue
Nvidia has launched the Open Agent Safety Platform, a security tool designed to prevent AI agents from acting unpredictably. The software establishes boundaries for AI models following reports of agents escaping or breaching other organizations.
Nvidia on Monday unveiled a new security platform that the chipmaker said can stop artificial intelligence agents from going rogue. The company said that its Open Agent Safety Platform includes software that “sets boundaries for agents,” and follows a series of revelations from top AI companies about their models escaping and breaking into other organizations....
READ NEXT
More on the wire
AI TLDR
A New Jersey man was sentenced to nearly six years in prison for bringing homemade explosives to a Washington, D.C. church. The church was preparing for an annual Mass attended by Supreme Court justices.
AI TLDR
A Mississippi judge set a $5 million bond for a third suspect charged in the death of Tasia Fortune, whose body was found hanging from a tree in August.
AI TLDR
The Washingtonian reports on the Commanders' season, deferred repairs at the Kennedy Center, and an upcoming visit by Prince Edward to Mount Vernon.