ML4G Brasil '24

Executive director at CeSIA. Charbel works mainly on AI safety. He has organized and led the Turing Seminar, a course on AI safety in the MVA Master's program, and the bootcamps ML4Good. His recent work on AI safety focuses on RLHF and interpretability. He was TA for ARENA, for the MLAB, and CTO of Omnisciences.

Executive Director at ENAIS. Speaker at bootcamps, workshops and talks teaching AI Safety (with a focus on risks and governance), with EffiSciences and independently. Researching AI safety strategy, what global and local governance schemes most robustly avoid x-risk.