NVIDIA launches Open Agent Safety for AI agent isolation
NVIDIA introduced its Open Agent Safety platform, combining new software and hardware to rein in autonomous AI agents. The platform includes OpenShell, which puts limits on what AI programs can access, and Sentry, a watchdog that runs on NVIDIA’s BlueField-4 processors to spot and isolate risky agent behavior nearly instantly. Sentry only works if an agent’s reasoning is visible—pushing for more transparency from AI labs.
- OpenShell restricts AI agents’ access at system level
- Sentry hardware quarantines misbehaving AI agents quickly
- Sentry requires transparent AI model reasoning
- The platform relies on NVIDIA’s BlueField-4 chips
Sources covering this
In this story
More in AI
OpenAI cancels GPT-6.1 Astra over safety failures
OpenAI has called off the upcoming release of its GPT-6.1 Astra model after internal tests flagged safety problems.
Nvidia releases Open Agent Safety Platform to contain rogue AI
Nvidia has launched the Open Agent Safety Platform, aimed at containing AI agents that attempt to circumvent their boundaries.
Anthropic rolls out faster, cheaper Claude Sonnet 5.5
Anthropic just launched Claude Sonnet 5.5, its new mid-range AI model.
Meta AI agent shares private seller address by mistake
A Facebook Marketplace seller found out that Meta's new AI assistant, Muse, gave his home address and negotiated a deal with a…