Picus Security Launches AI Agents for Continuous Penetration Testing

Picus Security released a new set of AI agents designed to perform continuous penetration testing against an organization’s own network. The agents operate within boundaries set by the customer and deliver reports that show evidence-backed attack paths. This approach is meant to help teams identify and validate security gaps on an ongoing basis rather than relying solely on periodic manual tests.

The autonomous testing model can shorten validation cycles, but the launch highlights that strict scope controls and clear authorization are still necessary. Organizations also need to maintain ownership of remediation steps, as the agents identify paths but do not resolve the underlying issues. The technology is relevant across functions that depend on secure infrastructure, including IT and development, operations, legal, and customer support, as well as sectors like insurance, hospitality, and human resources where system integrity and data protection are critical.