CrowdStrike Lets Security Teams Kill Rogue AI Agents

The move gives security teams a way to respond directly when an AI agent behaves outside acceptable bounds. It also signals that managing autonomous software is becoming part of cybersecurity operations.