I am currently the executive director at CeSIA.
I work mainly on AI safety. I have organized and led the Turing Seminar, a course on AI safety in the MVA Master's program, and the bootcamps ML4Good. You can find here my recent work on AI safety, which focuses on RLHF and interpretability.
I was TA in London for ARENA, in Berkeley for the MLAB, and CTO of Omnisciences. I've researched at Inria Parietal and Neurospin, and enjoy philosophy and jazz piano.
Here is my LinkedIn, Twitter, and Github.
Charbel-Raphael Segerie is the executive director of CeSIA, the French Center for AI Safety, and he is an OECD AI expert. He was previously Head of AI at EffiSciences, and founded ML4Good, an AI safety bootcamp designed for researchers, which has now been reproduced 15 times worldwide. He created the curriculum for an AI safety course at ENS, which was the first university-accredited course in the EU on general-purpose AI safety. He has experience in the industry as a startup CTO and has worked in different French research institutions (Inria, Neurospin). He was the capstone teacher at Arena and taught AI Safety in Berkeley for the MLAB. His research focuses on identifying emerging risks in AI, and he has published influential pieces on the limitations of current safety methods, such as RLHF and mechanistic interpretability and AI risk evaluations. He contributed to the EU AI Office’s Code of Practice for general-purpose AI systems.