Comment by Joe Kwon

Both separately require large frontier developers to address internal use through multiple mechanisms: frontier AI frameworks must describe how developers assess catastrophic risk “resulting from the internal use of its frontier models, including risks resulting from a frontier model circumventing oversight mechanisms” (SB 53 Section 22757.12(a)(10); RAISE Section 1421(a)(10)); assessments must be reviewed “as part of the decision to deploy a frontier model or use it extensively internally” (SB 53 Section 22757.12(a)(4); RAISE Section 1421(a)(4)).