Foresight's AI Salon: Samuel Nellessen | Slingshot: Automating Agent-to-Agent Jailbreaks with RL
Join Foresight Institute and Radboud University's Samuel Nellessen in his session entitled Slingshot: Automating Agent-to-Agent Jailbreaks with RL
Schedule
6 - 7 pm: Drinks & Social
7 - 8 pm: Talk, Q&A, & Breakouts
8 - 9 pm: Mingle & Hang
Speaker Bio: Samuel is an AI researcher specializing in empirical alignment and agentic security. His work focuses on transitioning safety evaluations from subjective NLP tasks into objective control problems. He is the creator of Slingshot, a reinforcement learning framework that autonomously discovers "Tag-Along Attacks" in tool-augmented LLMs, achieving zero-shot transferability across closed-source models like Gemini 2.5 Flash.
Salon discussions are held under Chatham House Rule (don’t connect people to ideas when discussing them outside of this workshop).
Foresight’s AI Nodes offer grant funding, local compute and community hubs for AI for Science and Safety projects in Berlin and the Bay Area.