Guidelines to Guardrails: Governing Agentic AI in Practice

Generative AI, retrieval-augmented generation AI, and agentic AI systems introduce new privacy and security threats that require novel and innovative approaches to governance and risk management. This session will cover what is different about these risks; explore real-world data flows and architectures for complex agentic AI systems; and discuss strategies to manage agentic AI risks for complex systems in production. Attendees will learn about emerging risk landscapes, legal requirements, and the latest governance techniques for mitigating these risks to evaluate and manage agentic AI systems safely and securely.

What you will learn: 

• Data flows and architecture of a complex RAG-enabled and agentic AI system. 
• Specific privacy, security, and other risks and controls for agentic AI systems.
• How to safely and securely move from proof-of-concept to production.