CyberArk (NASDAQ: CYBR), the global leader in identity security, has unveiled a significant innovation designed to secure the future of artificial intelligence in enterprise environments. With the launch of the CyberArk Secure AI Agents Solution, the company is responding to the exponential growth of AI agents—autonomous programs capable of executing complex tasks with little to no human oversight—and the profound identity-centric risks they introduce.

The solution is a timely response to predictions from Gartner, which estimates that by 2028, 25% of enterprise breaches will be caused by AI agent abuse, whether from external attacks or malicious insiders. These agents, which function with human-like autonomy but can scale like machines, represent a new class of digital identity that conventional identity and access management tools struggle to secure. CyberArk’s latest solution will be part of its Identity Security Platform, which is built to manage identities across diverse IT environments. It will enable organizations to implement identity-first security for agentic AI, treating each AI agent as a privileged identity subject to oversight, threat detection, and dynamic privilege control.

Future-Proofing Enterprise Security

Matt Cohen, CEO of CyberArk, emphasized the critical need for this advancement, stating, “When millions of autonomous, adaptable, and interactive AI agents gain privileged access to resources and services, organizations must not find themselves in a situation where security has lagged innovation.”

The Secure AI Agents Solution will bring the following key capabilities to enterprises:

• Discovery and Contextual Observability: Real-time visibility into both known and shadow AI agents across environments such as SaaS apps and AI infrastructures.

• Privileged Access Management: Enforcing least privilege, managing credentials, secrets, and certificates used by AI agents.

• Threat Detection and Response: Monitoring real-time behaviors to detect drift, anomalies, and potential misuse of privilege.

• Automated Lifecycle Management: Efficient onboarding, offboarding, and rightsizing of agent access throughout their lifecycle.

• Governance and Compliance: Ensuring AI agents operate in adherence to internal policies and regulatory standards.

These capabilities aim to match the agility of modern AI systems while maintaining strict control over access and behavior, preventing unauthorized use of data, infrastructure, and sensitive systems.

Developer Tools and Open Source Push

In parallel, CyberArk has introduced a new open-source security toolkit on GitHub, developed by CyberArk Labs. This toolkit enables developers to visualize how AI agents interact within systems, identify communication risks, and implement just-in-time credential provisioning. By providing these resources publicly, CyberArk supports a broader ecosystem of AI developers striving to build secure-by-design systems.

CORA AI – AI Within the Platform

CyberArk’s identity platform also integrates CORA AI™, its embedded AI engine. CORA AI goes beyond merely securing AI—it uses AI to enhance the platform’s own security functions. It continuously analyzes user and agent behavior, identifies emerging threats, provides automated security recommendations, and allows administrators to interact with the system through natural language commands, simplifying workflows and enhancing response times. This announcement builds upon CyberArk’s vision to integrate zero trust and intelligent privilege across all identities—human or machine. The company expects the full capabilities of the Secure AI Agents Solution to be available to customers by late 2025.

Strategic Outlook and Global Implications

CyberArk's approach comes at a crucial time as enterprise AI adoption increases and security threats become more complex. The inclusion of AI agents in critical processes—ranging from data processing to infrastructure control—requires organizations to rethink how digital identities are defined, secured, and governed. The announcement is also accompanied by cautionary guidance on the risks of delayed innovation in cybersecurity. Without integrated protection strategies, companies risk exposing themselves to breaches from AI agents with unsupervised privileges. CyberArk’s new offering addresses this gap directly by providing a scalable, automated, and intelligent defense layer specifically tailored for the AI-driven future.

For more technical resources, whitepapers, and product documentation, developers and security professionals can visit:

• CyberArk Secure AI Agents Solution

• CyberArk GitHub – AI Agent Toolset

• Blog: Securing Identities in the Agentic AI Landscape

As AI continues to reshape enterprise systems globally, CyberArk’s new initiative ensures that security does not become a bottleneck but rather a strategic enabler of innovation, trust, and resilience.