Securing Agentic AI: Identity as the Emerging Foundation for Defense

As enterprises continue deploying autonomous AI agents at scale, cybersecurity leaders are confronting new security and governance challenges. Traditional security controls for managing human and machine identities need to be enhanced to govern autonomous, decision-making systems that operate with elevated permissions and access. Identity is increasingly emerging as the control plane for securing these new digital workers.

This paper draws on multiple sources including a survey of 100+ CISOs across enterprise financial services and software companies and insights from cybersecurity, technology, and AI leaders.1 It defines agentic AI, examines how adoption is unfolding, outlines the emerging types of agents, the control frameworks needed to manage them safely, and the evolving role of identity security for agentic AI.