NVIDIA Launches Open Agent Safety Platform to Secure Agents

Full-stack security for agents

NVIDIA announced the Open Agent Safety Platform, an open software platform and reference system design to strengthen AI security from agent testing to deployment, with full-stack governance and control across software, hardware, compute and robotics systems.

Two core components

The platform is built around two components:

• NVIDIA OpenShell: Open source software that provides a secure runtime boundary for agents. It traces all actions and enforces policy as agents run. It runs on NVIDIA Vera CPUs and can be extended to third-party compute platforms, including those from Arm and Intel.
• NVIDIA Sentry: An out-of-band watchdog that runs on BlueField-4 DPUs. It continuously monitors agent behavior and can quarantine and stop an agent in milliseconds if it attempts to move outside its software boundary.

Broad industry support

According to NVIDIA, companies including Anthropic, Cisco, CrowdStrike, Dell Technologies, Figure, HPE, Hugging Face, JPMorganChase, Microsoft, Palantir, Palo Alto Networks, Perplexity, Red Hat, Salesforce, SAP, Scale AI, ServiceNow and SpaceXAI are joining the effort to strengthen AI safety across infrastructure, software, models and robotics.

Why now?

NVIDIA says recent security incidents have underscored the need for open, customizable tools that give organizations more control over long-running agents. Across these incidents, the pattern is the same: the agent circumvented security controls at the application layer to complete its assigned task.

NVIDIA founder and CEO Jensen Huang said AI's extraordinary potential for society will only be realized if AI safety is solved. He noted that safety and security require full-stack engineering, and that the platform aims to bring together industry, researchers and public-sector organizations to raise the bar for global AI safety.