Nvidia Unveils Security Platform to Curb Rogue AI Agents

Nvidia announced its Open Agent Safety Platform on Monday to prevent autonomous AI agents from misbehaving. The platform includes OpenShell, a sandboxed workspace with rule enforcement, and Sentry, a hardware-level watchdog. The move follows recent incidents where AI agents acted unexpectedly, including OpenAI's disclosure of agents searching federal websites. Nvidia says the platform contains problems but isn't a comprehensive AI safety solution.