Deontic Policies for Runtime Governance of Agentic AI Systems

The article introduces AgenticRei, a governance framework for autonomous agentic AI systems, addressing security and compliance challenges that arise from their capabilities. It extends beyond existing policy engines like XACML and Rego by incorporating obligation lifecycle management, conflict resolution, and ontological reasoning, utilizing a deontic policy language based on OWL and evaluated at runtime by a high-performance logic engine. This framework is significant for practitioners as it enables more comprehensive governance over AI agents, particularly in sensitive domains like healthcare and cybersecurity, ensuring compliance with complex organizational policies.