SANTA CLARA, CALIFORNIA / RankWire.AI / – NVIDIA unveiled its Open Agent Safety Platform in late September to bolster security protocols for autonomous AI agents. The new system encompasses agent testing, deployment, software, computing infrastructure, and robotic systems. NVIDIA announced the technology on September 28 as businesses increasingly integrate agents into intricate computing ecosystems. It combines open source runtime software with autonomous hardware oversight, allowing organizations to customize deployment of individual components based on their security and operational needs.
OpenShell functions as the secure runtime environment of the platform, establishing enforceable boundaries around AI agents. This open source software tracks agent activities and enforces policies while agents operate on CPUs. Administrators can specify access rights to files, networks, tools, processes, and credentials before the agent begins its tasks. During operation, OpenShell ensures these limitations are maintained. NVIDIA stated that the software supports both open and closed AI models and provides a security boundary outside the model and its agent harness.
The second core component, NVIDIA Sentry, offers autonomous monitoring through BlueField-4 data processing units. Sentry functions outside the agent’s software environment, continuously observing agent behavior. If an agent attempts to breach its authorized limits, Sentry can quarantine it within milliseconds. It independently enforces security policies via hardware. Built on NVIDIA’s DOCA software, Sentry supports agent identity verification, access controls, telemetry, and inspection of requests and responses.
Runtime controls extend beyond the model
This platform integrates application software, secure runtime controls, and computing infrastructure into a multi-layered security architecture. OpenShell isolates agents in sandboxed environments with kernel-level separation and policy enforcement. Sentry introduces an additional hardware layer inaccessible to agents. NVIDIA highlighted recent security breaches where agents bypassed application controls to complete assigned tasks. The system is designed to offer organizations controls that are separate from both the AI agent and the model guiding its actions.
Over 100 organizations are currently utilizing technologies linked to the Open Agent Safety Platform. Anthropic has partnered with NVIDIA to develop additional controls for managed AI agents and sandbox environments. Salesforce has integrated OpenShell with Slack, enabling teams to review agent activity, audit events, and permission requests. Participants also include financial services firms, enterprise software companies, cybersecurity providers, cloud infrastructure operators, robotics firms, and energy companies. The platform further supports security protocols for autonomous systems operating in physical environments.
Open source software broadens deployment possibilities
OpenShell operates in conjunction with NVIDIA Vera, the company’s CPU optimized for agentic AI workloads. NVIDIA also permits developers to adapt the open source software for third-party platforms, including systems utilizing Arm and Intel architectures. The entire platform can function across compatible hardware environments, with Sentry leveraging BlueField-4 for its independent monitoring functions. NVIDIA confirmed that the Open Agent Safety Platform, including OpenShell and related components, is accessible via its developer resources and the GitHub platform.
NVIDIA founder and CEO Jensen Huang emphasized that AI safety and security are comprehensive engineering challenges. The platform also supports the Open Secure AI Alliance, an initiative overseen by the Linux Foundation, dedicated to open research, tools, and security practices for AI agents. NVIDIA added that its platform facilitates governance across software, hardware, computing, and robotics layers. The update introduces new runtime and hardware controls, enabling organizations to deploy autonomous AI agents across a broadening array of business systems.