Comment by Michael Bennet

AI systems should undergo regular public risk assessments to examine their safety, reliability, security, explainability, and efficacy. Unverified source (2023)