Exclusive | Fired OpenAI Researchers Ask Company to Preserve Visibility Into AI Reasoning - WSJ

In the letter, the signatories called on the company to work with outside safety auditors. They also asked OpenAI to preserve the ability to monitor increasingly sophisticated AI models.

The letter states that the three former employees were concerned that AI companies could lose the ability to monitor AI systems' chain-of-thought. The report describes the signatories as three people who were fired by OpenAI, and identifies the recipients as the company's board members and its safety committees. The excerpt published with the article does not name the three individuals. It also does not include a response from OpenAI. No details of the letter beyond those requests are included in the excerpt.

The two requests in the letter concern external review and internal monitoring. On external review, the signatories asked OpenAI to work with outside safety auditors. On internal monitoring, they asked the company to preserve visibility into how increasingly sophisticated AI models reason. According to the excerpt, the signatories wrote that they were concerned AI companies could lose the ability to monitor AI systems' chain-of-thought.