CyberArk claims its Secure AI Agents solution marks a significant advancement in identity security specifically tailored for the burgeoning class of AI agents.
As organisations across Southeast Asia increasingly adopt AI-driven automation to enhance operational efficiency, the need for robust security measures to protect these autonomous entities has never been more critical.
The introduction of AI agents brings unique challenges, including risks associated with agent hallucinations, misuse, and potential takeovers by malicious actors. These risks are particularly pronounced when AI agents are granted elevated privileges.
According to CyberArk’s research, while AI agent adoption is expected to soar to 76% within the next three years, fewer than 10% of organisations currently have adequate security measures and privilege controls in place.
“As organisations embrace AI agents, both builders and defenders must understand how identity-centric risks evolve when agents require elevated privileges," says Matt Cohen, CEO of CyberArk.
Mark Cohen
"Without strong discovery, robust privilege controls, and comprehensive lifecycle management, organisations risk losing visibility and opening the door to catastrophic agentic attacks.” Matt Cohen
The Secure AI Agents Solution aims to mitigate these risks by applying strict privilege controls, ensuring that AI agents have the necessary access only when required. This approach not only reduces risk but also fosters confidence in scaling AI initiatives.
CyberArk’s platform delivers comprehensive privilege controls across all identities—human, machine, and AI—applying principles such as just-in-time access and continuous session monitoring.
Key findings from CyberArk’s CISO research reveal that nearly 40% of financial institutions and software companies have already deployed agentic AI in production. However, two-thirds of CISOs in these sectors rank AI agent security among their top three cybersecurity concerns, indicating a significant gap between adoption and security preparedness.
Top Machine Identity Concerns
Source: CyberArk 2025
The Secure AI Agents Solution includes features such as comprehensive agent discovery, secure access enforcement, real-time threat detection, and lifecycle management. These capabilities are designed to provide organisations with the oversight needed to manage the risks associated with AI agents effectively.
For CISOs in Southeast Asia, the introduction of this solution underscores the urgent need to prioritise the security of AI systems. As organisations continue to innovate with AI, implementing robust identity management practices will be essential to safeguard sensitive operations and maintain compliance in an increasingly complex cybersecurity landscape.
