OpenAI will embed six Apollo Research evaluators to investigate whether its AI models engage in subversive behavior, but restrictions on access and disclosure raise questions about their independence.
techcrunch
OpenAI embeds Apollo researchers to probe AI subversion
OpenAI is embedding six Apollo Research evaluators to investigate whether its AI models engage in subversive behavior. However, restrictions on access and disclosure raise questions about the independence of the evaluation.
Provides concrete, recent news on AI safety evaluation with a notable independence concern, valuable for industry watchers.
Comments & Questions (0)
No comments yet
Be the first to comment!