OpenAI's big idea to increase the safety of its tech is to have AI models police each other

• OpenAI is experimenting with a technique to enhance transparency with its AI models.
• The method involves powerful AI models explaining their thought processes to a second AI.
• This initiative follows significant changes in OpenAI's safety department earlier this year.

OpenAI has a new technique for getting AI models to be more transparent about their thought processes: Getting them to talk to each other.

The company showcased the research behind the technique this week and will unveil more details in a forthcoming paper, according to Wired.