About this talk
This talk explores the security implications of AI agents with production access, focusing on the need for careful management of their capabilities. The speaker presents a framework for identifying dangerous combinations of AI tools, emphasizes the importance of implementing layered defenses through tool proxies and detection mechanisms, and discusses strategies for reducing the blast radius in the event of control failures. Attendees will gain insights into navigating the complexities of deploying AI technologies securely in production environments.
More from this event
See all 91 talks →
BSidesSF 2026 - Opening Remarks (Sunday) (Reed Loden)
14:36
BSidesSF 2026 - Follow the data to learn the secret (Dylan Ayrey)
35:17
BSidesSF 2026 - Not My Vibe: When AI Coding Agents Go Off the Rails (Aonan Guan, Zhengyu Liu)
45:56
BSidesSF 2026 - "Ask the EFF" Panel (Panel)
45:30