Center for AI Safety (CAIS) — Description: The Center for AI Safety (CAIS) is a San Francisco-based nonprofit focused on reducing societal-scale risks from AI through technical safety research, field-building, and public communication. Founded by Dan Hendrycks and Oliver Zhang. Known for the MMLU benchmark, representation engineering, and the May 2023 "Statement on AI Risk" signed by 350+ AI leaders.

NoteThe claim makes four main assertions: (1) San Francisco-based nonprofit focused on reducing societal-scale risks through technical safety research, field-building, and public communication — CONFIRMED by source (describes research, field-building, and advocacy). (2) Founded by Dan Hendrycks and Oliver Zhang — CONFIRMED (both listed in Leadership section). (3) Known for MMLU benchmark — UNVERIFIABLE (not mentioned in source). (4) Known for representation engineering — UNVERIFIABLE (not mentioned in source). (5) May 2023 'Statement on AI Risk' signed by 350+ AI leaders — CONTRADICTED/OUTDATED: Source states 'Global Statement on AI Risk signed by 600 leading AI researchers and public figures' with no date specified. The claim cites 350+ signers from May 2023, but the source shows 600 signers (a larger number, suggesting the statement may have grown or the source reflects updated data). This is either outdated information or a different statement. Overall, the core description is confirmed, but specific achievements (MMLU, representation engineering) are unverifiable, and the statement signature count/date is inconsistent.