Content

Speaker

Andy Zou

Abstract

This presentation demonstrates how red teaming uncovers critical vulnerabilities in AI agents that challenge assumptions about safe deployment. The talk discusses the risks of integrating AI into real-world applications and recommends practical safeguards to enhance resilience and ensure dependable deployment in high-risk settings.

Bio

Andy Zou is a PhD student at CMU. He is the CTO and cofounder at Gray Swan AI and a cofounder of Center for AI Safety. He works in AI Safety and Security.

Host

UMass AI Security