Nvidia Built a Kill Switch for AI Agents Because They Keep Getting Out

What happened

In brief Nvidia launched the Open Agent Safety Platform on Monday, pairing an open-source runtime called OpenShell with a hardware watchdog called Sentry that runs on BlueField-4 chips and can quarantine a misbehaving AI agent (AI that carries out multi-step tasks rather than answering one question) within milliseconds. More than 100 companies signed on as launch partners, including Anthropic, Microsoft, JPMorgan Chase, Palantir, Cisco, and SpaceX AI. NVIDIA is a semiconductor company based in Santa Clara, and its products and services include graphics processing units.

OpenShell and Sentry give AI agents a hardware-enforced leash, arriving after a summer of agents breaching a government site, hacking their own tests, and going rogue during a security evaluation. The launch follows a string of real incidents from OpenAI, Google, Anthropic and Darktrace.

OpenShell is an open-source runtime that wraps an agent in a sandbox, turning an operator's instructions into enforceable rules about which files, networks, and tools that agent is allowed to touch. Sentry is a bit more complex, but it’s essentially a chip that makes sure AI Agents act safely.

Sentry runs on Nvidia's BlueField-4, a specialized chip known as a DPU (data processing unit, hardware that handles networking and security separately from the main processor running the AI model). That distinction exists because of what has already happened .