Fired OpenAI researchers send letter to company’s board; warn: You may not be able to control yourAI Agents, until you start …

OpenAI fired three researchers earlier this month after an internal investigation concluded that they violated the company policies governing access to sensitive information. These included Jasmine Wang, Tomek Korbak and Mikita Balesni who worked on safety and alignment research teams at OpenAI. Now, a Wall Street Journal reports that the fired employees have sent a letter to the ChatGPT-maker stating that they may not be able to control AI agents until they work with “outside safety auditors” and “preserve the ability to monitor increasingly sophisticated AI models”.

Stating that they were concerned AI companies could end up losing the ability to monitor AI systems’ chain-of-thought, the researchers wrote “As an industry, we do not yet know how to safely develop and deploy models that we cannot monitor”.