OpenAI Outlines Framework for Independent Model Evaluations
OpenAI shares lessons on designing trustworthy third-party evaluations for frontier AI models, emphasizing the role of task environments and validity checks.
OpenAI shares lessons on designing trustworthy third-party evaluations for frontier AI models, emphasizing the role of task environments and validity checks.
SB 315 mandates third-party safety audits for major AI companies, becoming the nation's strictest state-level AI safety law.