Zhiyuan FlagSafe platform officially released

Recently, Beijing Zhiyuan Artificial Intelligence Research Institute, together with Peking University, Beijing University of Posts and Telecommunications, Beijing University of Aeronautics and Astronautics, Shanghai Jiaotong University, Information Engineering Institute of the Chinese Academy of Sciences, Institute of Computing Technology of the Chinese Academy of Sciences and other domestic institutions, officially released the FlagSafe big model security platform. The platform is the first to gather multiple cutting-edge large-scale model security research projects, focusing on three core directions: red team exercises, blue team defense, and white box perspective, to jointly create a high standard security platform covering risk discovery, defense governance, and mechanism interpretation.