Your AI Agents Are Overprivileged: The Case for Fine-Grained Authorization

Generative AI is booming, but most enterprises still rely on coarse access controls, leaving RAG and agentic systems free to roam confidential data. In this NDC Copenhagen talk, Ashish Jha makes the case for Fine-Grained Authorization (FGA) to enforce context- and document-level permissions at scale—no more overprivileged bots.

You’ll see real-world demos with OpenFGA and LangChain covering multi-tenant isolation, leakage prevention, audit trails, and handling billions of access checks without slowing down. If you’re building internal copilots or customer-facing AI, this is the security upgrade you can’t skip.

Watch on YouTube