Comment by Calogero Bono

Europe cannot limit itself to regulation. It must develop its own AI safety expertise, fund open-source red-teaming projects, and demand that internal security tools be documented and verifiable by third parties.