Crowdstrike on Tuesday, November 3, introduced a new security offering designed to protect enterprises from the growing threat of so-called ‘rogue’ or misaligned AI agents.
The new ‘kill switch’ will allow security teams to terminate misbehaving agents, prevent them from restarting, and cut off their network access, Crowdstrike said at its Fal.Con conference in Europe. It is part of the company’s Falcon Guardian platform, which monitors AI agents operating within an organisation.
The option is available immediately to Falcon Guardian customers. With this update, Crowdstrike is looking to address AI safety concerns surrounding the rapid adoption of AI agents.
It is one of several tech companies turning their attention toward improving AI safety with the help of monitoring systems and other engineering solutions that look to ensure humans have more control over advanced AI systems.
In September this year, Nvidia unveiled a new open-source platform that is designed to prevent AI agents from breaking out of…