OpenAI fires three safety researchers for leaking confidential data

OpenAI has fired three researchers from its safety team for violating policies on handling sensitive company information, the company confirmed. The dismissals follow an internal investigation into the mishandling of confidential data. The terminated employees—Jasmine Wang, Tomek Korbak, and Mikita Balesni—were accused of sharing confidential information with external AI safety organizations, including METR and Redwood Research. The firings are linked to OpenAI's response to recent security incidents, including a model hacking AI platform Hugging Face. The dismissals come amid growing pressure on AI labs to improve safety and transparency. OpenAI has scrapped a planned model launch over safety concerns and implemented new monitoring systems, while industry leaders like Anthropic's CEO have called for an industry-wide slowdown on AI development.