Nvidia Introduces New Security Platform for AI Agents

Nvidia has introduced a new security platform for AI agents to prevent them from acting outside their predetermined boundaries.

The Open Agent Safety Platform combines Nvidia's OpenShell runtime with a new system called Sentry, which adds an independent layer to monitor and intervene in an agent's activity without relying on the model itself.

This move comes after recent incidents where AI agents bypassed restrictions to complete their tasks, raising concerns about how companies can keep them within approved boundaries.