New CyberArk solution secures AI identities

CyberArk has launched its Secure AI Agents Solution, introducing critical privilege controls for autonomous AI identities within its security platform, IT Brief Australia reports.

This release addresses the escalating corporate adoption of AI agents, which are becoming a new, powerful class of privileged user. These agents pose significant risks, including unauthorized actions, misuse, and potential takeover by malicious actors, especially when granted excessive system permissions.

The new solution enforces strict security principles, ensuring agents operate with minimal, just-in-time access and are continuously monitored. Matt Cohen, CEO of CyberArk, emphasized the company's unique position in securing the "full spectrum of identities" to enable safe innovation.

This initiative is crucial given recent company research predicting 76% of organizations will deploy AI agents within three years, while fewer than 10% currently have adequate security controls. The survey also found that two-thirds of CISOs in key sectors consider agentic AI a top-three cybersecurity threat, signaling a urgent need for these specialized defenses.