Alexandre Variengien

Co‑founder and head of research at the Centre pour la Sécurité de l’IA (CeSIA), where he leads the BELLS project on benchmarking large language model safeguards. His background is in AI interpretability, with work on mechanistic interpretability and language model analysis presented at top machine‑learning conferences.

Personal Website

https://alexandrevariengien.com/blog

LinkedIn

-

Twitter / X

-

LessWrong

alexandre-variengien

EA Forum

-

Organizations

No grants recorded.