My work on "Frontier AI Regulation" last year was based on this policy red-teaming approach 🙂

→ View original post on X — @jeremyphoward