Can AI safety frameworks keep pace with frontier model risks? - Lexology

Can AI safety frameworks keep pace with frontier model risks?  Lexology