Delinea Introduces Runtime Authorization for AI Agents

Delinea said it released runtime authorization capabilities for AI agents, aimed at enforcing what agents do after they gain access. The company focused on policy and recording for actions within a session, rather than only controlling access when connections start.

Delinea said many identity security approaches verify connections when a session opens and then audit or revoke access after the fact, or rely on visibility limited to traffic routed through their own server. It described this as leaving gaps for direct connections outside that view, including cases where an agent can perform actions before controls are applied action by action.

The runtime authorization capabilities are described as using a single control point to authorize and record agent connections across databases, SSH hosts, Kubernetes clusters, cloud consoles, and MCP servers. Delinea said the platform injects credentials just-in-time at the moment of access, scopes them to the task, and revokes them when the task completes, then evaluates tool calls individually for allow, block, or pulling in a human before execution.

Art Gilliland, CEO at Delinea, said “The security industry spent years solving credential theft, but AI agents have introduced a new problem: authorized access doing unauthorized things,” adding, “The perimeter has moved to inside the session, and no one has figured out how to address that until now.” Dhruv Kumar of Telnyx said, “With Delinea, every action is now tied to a named identity and recorded at the action level.” Delinea said the capabilities were made available as part of the Delinea Platform.

Provided by Globe Newswire on behalf of Delinea. Click to read original content. The original article was written by Decision Insights Editorial.