A Case Study of Vulnerabilities in AI Agents

This talk will present Jack's original security research into vulnerabilities in AI agents. Spanning prompt injections, broken authentication, and more, this will feature some of the most critical vulnerabilities he has found in the wild – affecting widely used applications and open source repositories with tens of thousands of stars. He will also discuss how we can more securely architect AI agents going forward to prevent these and other vulnerabilities.

Save your spot for this session!