AI Agents and Their Impact on Governance and Security

This report examines the evolving landscape of AI agents and their implications for governance, risk management, and digital trust. It identifies the challenges organizations face in securing access for both human and non-human identities, particularly as the deployment of AI agents accelerates. The report categorizes AI agents based on their level of autonomy and associated risks, detailing the security vulnerabilities introduced by assisted, low-code, and fully autonomous agents. It highlights the urgent need for organizations to develop governance policies, as 57% are still in the planning stages. The report presents survey data from security and identity leaders, revealing that 91% of organizations are deploying or evaluating AI agents, emphasizing the immediate need for effective governance frameworks. Key challenges include visibility gaps, excessive access permissions, and compliance risks. The report advocates for an identity-centric approach to manage these risks and ensure secure innovation in the agentic era.