Securonix Adds Behavioral Analytics for Human and AI Agent Threats

Securonix has introduced Advanced Behavioral Analytics, a capability designed to identify activity that may be technically permitted but operationally abnormal across users, systems and AI agents.

The product combines user and entity behavior analytics with new agent-focused baselines. It is intended to flag cases such as an enterprise agent using an unexpected tool, reaching data outside its purpose or changing its operating pattern. The system connects those deviations with identity, authority, asset sensitivity and timing.

Securonix says consequential response actions remain subject to human approval, with policy and audit records carried through investigation and case management. The company also announced ISO/IEC 42001:2023 certification for its AI management system.

As enterprises give agents credentials and tool access, conventional identity alerts may miss misuse that occurs within granted permissions. Behavioral controls can add context, but their value will depend on transparent baselines, manageable false positives and clear limits on automated response.