OpenAI will embed safety researchers to investigate AI subversion
Originally published by TechCrunch

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.
Original reporting
Read the full story at the original publisher
PulsePress is a news discovery platform. We bring headlines from multiple publishers together so you can discover stories in one place. The complete reporting remains with the original publisher.
Read original article →