As organizations increasingly adopt artificial intelligence (AI) agents, they face new challenges in defining the conditions under which these agents are authorized to act, and in ensuring this authority is enforced as systems evolve. Many agents in a portfolio may share the same foundational model, which can lead to systemic vulnerabilities across the entire agent estate, necessitating authorization and monitoring in each instance.
Developed in collaboration with Capgemini, AI Agents in Action: A Playbook for Trusted Adoption, Authorization and Scaling introduces the Agent Capability and Authorization Profile (ACAP) as a deployment-level governance and authorization instrument for agentic systems. Encompassing delegation policy, system design and operational oversight in a single framework, the ACAP makes delegated decisions and actions auditable, enforceable and accountable across the full life cycle of deployment.
The report outlines clear agent guidelines and enforceable authorization profiles, providing organizations with a structured path from pilots to portfolios of governed agents at scale. It serves as a practical resource for teams seeking to operationalize AI agents responsibly while maintaining auditability, enforceability and human accountability and advancing the development of shared standards.