Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the critical security challenges and solutions for AI agents in enterprise environments through this 37-minute conference talk from AWS re:Invent 2025. Discover why traditional authentication and authorization models fall short when AI agents operate across multiple systems and data sources, and learn about the risks of unchecked agent behavior and compliance challenges. Examine emerging security approaches including persona shadowing, delegation chains, capability tokens, and human-in-the-loop escalation mechanisms. Gain insights into evolving standards such as OAuth, OIDC for Agents, GNAP, and verifiable credentials that are shaping the future of agent identity management. Understand practical frameworks for building safer, scalable agentic systems based on WorkOS's real-world experience with SaaS and AI companies. Learn how to implement first-class agent identities and develop robust identity and access management strategies for AI-driven enterprise software transformation.
