Evoke Security introduced its Rogue Agent Assessment on October 28, offering businesses a review of how artificial-intelligence agents operate across their systems.
The company said the assessment is available now and covers three areas: misuse of agents that bypasses security or usage rules, agent behavior that departs from instructions or creates unnecessary risk, and unapproved AI tools used within an organization.
Its approach examines current and historical runtime behavior, according to Evoke. The company described four questions for evaluating an action: whether it serves the original mission, whether the method is reasonable, what intent the behavior suggests and how much potential risk it creates.
The assessment is intended to help security teams identify existing agents, understand their behavior and decide where intervention is needed. Evoke said those decisions can include allowing an agent to continue, redirecting it or blocking it.
Those are the vendor’s descriptions of the service. The announcement does not establish independently measured detection accuracy or effectiveness across enterprise environments.
Sources Evoke Security — Product announcement: https://www.prnewswire.com/news-releases/evoke-security-introduces-first-of-its-kind-rogue-agent-assessment-to-surface-risky-ai-agent-behavior-inside-enterprises-302921948.html