Saviynt adds intent-aware runtime controls for AI agents

Saviynt has released major enhancements to its Agent Access Gateway, a runtime authorisation layer that controls what AI agents can do as they interact with applications, data, APIs, tools, infrastructure and other agents.

The headline addition is Intent-Aware Runtime Authorization (IARA), which evaluates an agent’s actions in real time based on identity, context, policy and intent, and can block an action and generate an audit event if it falls outside approved boundaries.

The release also adds inbound and outbound access controls that let organisations govern who can interact with an AI agent and what data and applications the agent can reach, and the ability to recognise whether an agent is acting independently, on behalf of a human, or via another agent.

Saviynt extended its native integrations to include Microsoft Foundry, n8n and Snowflake Cortex, among others.

Separately, Saviynt added identity-verification capabilities to its platform to counter impersonation risk, including biometric scanning, selfie and liveness detection and support for more than 4,000 government-issued document formats across over 177 countries.

“AI agents are becoming a new class of enterprise identity — autonomous, powerful, and capable of taking action across critical business systems,” said Vibhuti Sinha, chief product officer at Saviynt.

“Agent Access Gateway gives enterprises a way to control AI agent behaviour at runtime, when decisions actually happen. With IARA, organisations can move beyond static permissions and make access decisions based on what an agent is trying to do, why it is doing it, and whether that action should be allowed.”