Nvidia's AI Autonomy Controls Spark Governance Questions

Nvidia's latest move has sparked questions about who governs AI autonomy. The company launched its Open Agent Safety Platform, which offers enterprises a new layer of controls for governing agent access and actions.

The platform combines OpenShell, an open-source runtime that controls an agent's access and actions while it operates, with Sentry, a reference design for monitoring and enforcing those limits using a separate control layer. This approach reflects a shift in how organizations must govern AI systems as they become more autonomous.

Lee Rossey, CTO and co-founder of SimSpace, said that controls should operate independently of the agent itself as agents gain the ability to call tools, access systems, and make decisions autonomously. He emphasized that an agent can create a security problem without acting maliciously.