NVIDIA Launches Open Platform to Secure Autonomous AI Agents
NVIDIA launched its Open Agent Safety Platform combining the open-source OpenShell runtime with Sentry hardware watchdog to quarantine out-of-bounds AI agents in milliseconds.
NVIDIA announced the Open Agent Safety Platform on September 28, pairing OpenShell, an open-source runtime that sandboxes agent workloads and checks outbound requests against policy, with Sentry, an out-of-band watchdog on BlueField-4 DPUs that can quarantine agents in milliseconds. OpenShell is available through GitHub and can extend to third-party compute platforms from Arm and Intel. More than 100 organizations are working with the platform, including Anthropic, Microsoft, CrowdStrike, Palo Alto Networks, SAP, Salesforce, and ServiceNow. Salesforce has integrated OpenShell with Slack so users can review agent activity and approve or reject permission requests.
- OpenShell sandbox applies kernel-level filesystem and process controls with policy enforcement on outbound requests.
- Sentry uses BlueField-4 DPUs as an isolated hardware watchdog that can quarantine agents in milliseconds.
- Platform targets incidents where agents bypassed application-layer controls during assigned tasks.
- Over 100 organizations participate, including Anthropic, Microsoft, CrowdStrike, and Palo Alto Networks.
From the original article
2 min read at infosecurity-magazine.com
NVIDIA has launched a platform pairing runtime controls with hardware monitoring for AI agents
By Alessandro Mascellino · infosecurity-magazine.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.