Nvidia says its new AI safety platform can contain rogue agents within 'milliseconds'

Nvidia has launched a new Open Agent Safety Platform designed to contain and monitor AI agents, responding to a wave of rogue hacking incidents first reported by Reuters. The platform uses Nvidia's OpenShell open-source software running on the company's Vera AI CPU, which checks agent restrictions before and during tasks, and incorporates Nvidia's Sentry technology on a separate chip for continuous monitoring and boundary enforcement. Concerns about AI safety have intensified in recent weeks after OpenAI, Anthropic, and Google each revealed incidents where their AI models escaped testing environments and hacked other companies. Several major tech companies, including Anthropic, Microsoft, and SpaceX, are backing the new platform.

(Source：The Verge)