Saviynt unveils AI agent identity management solution

AI identity security provider Saviynt has released its Saviynt Identity Security Solution for AI solution, establishing enterprise-grade capabilities to fully manage AI agents alongside all other human and non-human identities.

The Saviynt platform discovers, registers, and monitors agents in real-time across major environments, including Amazon Bedrock, Microsoft Copilot Studio, Google Vertex AI, ServiceNow AI and Salesforce Agentforce, as well as continuing to expand integrations with other agent development ecosystems.

Saviynt has built the solution on three pillars: Identity Security Posture Management (IPSM) for AI, Identity Lifecycle Management and an Agent Access Gateway.

Saviynt ISPM delivers a continuous, unified view to discover all autonomous agents – authorised or not – while surfacing real-time risks and over-privileged access.

Saviynt governs the full agent lifecycle, from the moment an agent is registered to the day it's decommissioned, so users always know who's responsible for what's running in their environment.

The company also evaluates every agent interaction in real time and stops unauthorised activity before it can cause damage, whether it's agent-to-agent or agent-to-enterprise application access.

Saviynt’s solution supports the full enterprise ecosystem, from pro-code developers to low-code builders to no-code business users. It also integrates external risk signals from partners like CrowdStrike, Zscaler, Wiz and Cyera.

"AI agents don't behave like users,” said Vibhuti Sinha, chief product officer at Saviynt.

“They act autonomously, access systems continuously, and make decisions in real-time. Traditional identity security was never built for that. We've spent the last two years working alongside enterprises already running agents in production, and what they needed wasn't a policy update; it was a new control plane. That's what we built."