As part of our continued work to help assure the future security of deployed cryptographic systems, we recently released new code that will enable researchers to benchmark AI-based attacks — and compare them to new and existing attacks going forward

→ View original post on X — @aiatmeta,