Rubrik launches Agent Identity to secure autonomous AI agents

Rubrik has introduced Agent Identity, a new component of its Rubrik Agent Cloud platform, to protect systems from autonomous AI agents. The feature leverages the SAGE framework to perform semantic analysis of agent requests before sensitive actions are executed, allowing policies to be evaluated in context rather than relying solely on static identities or permissions. When a request conflicts with corporate security rules, Agent Identity can block the operation in advance.

The solution also includes a dynamic policy engine and the MCP Gateway, which centralizes control of resources and APIs used by agents. Rubrik aims to achieve “Agentic Cyber Resilience,” matching the speed of AI‑driven actions with real‑time monitoring and enforcement. According to Rubrik Zero Labs, 86 % of IT and security leaders believe AI agents will outpace existing protections within the next twelve months, and 88 % are concerned about meeting recovery objectives under this new threat landscape.