TLDR: CyberArk has announced its Secure AI Agents Solution, an industry-first offering designed to protect autonomous AI agent identities with comprehensive privilege controls. This new solution addresses the escalating security risks associated with AI agents requiring elevated access to perform automated tasks within enterprise environments.
NEWTON, Mass. & PETACH TIKVA, Israel – November 4, 2025 – CyberArk (NASDAQ: CYBR), a global leader in identity security, today announced the general availability of its Secure AI Agents Solution. This innovative offering expands the CyberArk Identity Security Platform by introducing purpose-built privilege controls to safeguard the rapidly growing class of AI agent identities. The solution is slated for general release in December 2025, with further enhancements planned for 2026.
The proliferation of AI agents, which automate tasks and enhance operational efficiency, has introduced a new and powerful class of privileged identities. These autonomous entities, however, present novel security risks, including agent hallucinations, potential misuse, and the threat of takeover by malicious actors. These risks are significantly amplified when AI agents operate with elevated privileges within an organization’s systems.
CyberArk’s Secure AI Agents Solution directly confronts these challenges by implementing stringent privilege controls. The core principle is to ensure that AI agents are granted only the necessary access, precisely when it is required, and nothing more. This strategic approach is designed to mitigate risk, prevent unauthorized access, and enable enterprises to confidently scale their AI-driven initiatives while maintaining robust security postures.
Matt Cohen, CEO of CyberArk, emphasized the critical need for such solutions, stating, “As organizations embrace AI agents, both builders and defenders must understand how identity-centric risks evolve when agents require elevated privileges. Without strong discovery, robust privilege controls, and comprehensive lifecycle management, organizations risk losing visibility and opening the door to catastrophic agentic attacks. CyberArk uniquely secures the full spectrum of identities — humans, machines and AI agents — by applying the right level of privilege controls, enabling innovation while maintaining security and compliance.”
Research conducted by CyberArk highlights the urgency of this new security paradigm. A study of 104 Chief Information Security Officers (CISOs) across North America and Europe revealed that AI agent adoption is projected to reach 76% within the next three years. Despite this rapid anticipated growth, fewer than 10% of organizations currently possess adequate security controls to manage these agents. Alarmingly, nearly 40% of enterprise financial institutions and software companies already have AI agents deployed in production environments. The research further indicated that two-thirds of CISOs in financial services and software sectors rank agentic AI among their top three cybersecurity concerns, with over one-third identifying it as their primary concern.
The Secure AI Agents Solution is engineered to deliver several key capabilities:
Comprehensive Agent Discovery: Automated detection of AI agents across SaaS, cloud, and developer environments, complete with enriched profiles detailing ownership, roles, and access rights.
Secure Agent Access: Enforcement of strong authentication and the principle of least privilege, incorporating zero standing privileges and thorough auditing of agent activities.
Real-Time Threat Detection: Continuous monitoring for anomalies and unauthorized access attempts, triggering automated alerts and facilitating rapid response.
Lifecycle Management: Support for compliance requirements throughout the entire lifecycle of AI agents.
Also Read:
- Zenity Unveils Real-Time Security for OpenAI AgentKit-Powered AI Agents
- Frontegg Launches AgentLink to Securely Integrate SaaS Applications with Agentic AI
By extending its proven identity security capabilities to autonomous AI agents, CyberArk aims to provide a unified, privilege-first approach that ensures every identity—human, machine, and AI—is governed, secured, and monitored with consistent rigor. This strategy is crucial for fostering innovation without compromising an organization’s security posture or regulatory compliance.
