BLISS AI Speaker Series: Safety and Security of LLM Reasoning — Jonas Geiping

A 45-minute research talk from the BLISS AI Speaker Series, with Jonas Geiping, Research Group Leader at the Max Planck Institute for Intelligent Systems, on the safety and security of LLM reasoning: where reasoning chains can be attacked, and what that means for systems already in production. Hosted by BLISS Berlin e.V. at TU Berlin's chemistry institute on Straße des 17. Juni. Talk first, then open discussion over free drinks and pizza. Aimed at students, PhD candidates, academic researchers and industry people doing ML research. RSVP on Luma is required for entry and doors close at 19:15. Photography may take place. Why go: Geiping's group publishes some of the sharpest adversarial work on language models in Europe, and this is the applied end of it. Best for people building or red-teaming LLM systems rather than following safety as a policy topic.