Abnormal AI Research Finds Security Teams Lack Baselines for AI Agents

Abnormal AI research finds nearly half of security teams say they cannot tell what normal behavior looks like for AI agents. The teams are prioritizing behavioral monitoring to catch unexpected activity.