OpenAI partners with 4 orgs for AI model safety tests
openai begins embedded assessments with apollo research, metr, transluce, and goodfire to evaluate model alignment. this affects openai's ai safety and governance.
openai has initiated a new phase of embedded assessments with four partner organizations: apollo research, metr, transluce, and goodfire. the goal of these assessments is to independently evaluate the alignment, safeguards, and monitoring systems of openai's models. by collaborating with external entities, openai is taking a proactive approach to ensuring the safety and reliability of its ai systems. this move signals openai's commitment to transparency and accountability in ai development. the outcomes of these assessments may influence openai's future ai safety strategies.