Authorization Masterclass: Policy-Based Access for AI Agents

This technical masterclass was designed for IAM architects and security engineers evaluating how existing authorization models apply to AI agents.

PlainID Field CTO Vladi Berger examined why agentic systems challenge traditional RBAC and how attribute-based, policy-driven authorization can define, enforce, and audit fine-grained access.

The session covered PBAC design patterns for agentic systems, a live demonstration of data-level authorization in an LLM pipeline, and Q&A with the PlainID engineering team.