Guardrails First: Building AI Agents That Won’t Leak Your Secrets - Track 2

Why watch

This OWASP talk covers implementing security guardrails in AI agents to prevent sensitive data and secrets from being exposed through AI interactions. Designed for developers and security engineers building AI-powered applications who need to understand practical patterns for securing AI agents against data leakage attacks.