OpenAI safety employee resigns, claiming the company’s ‘culture is broken’

A safety employee at a major AI company resigned and published an essay claiming the company's culture is broken. The employee, who led the writing of safety reports for major product launches, argued that the company's approach of testing systems and fixing problems after they occur creates dangerous risks as AI systems become more capable, and that frontier AI companies should instead operate with the careful planning and redundancy used in nuclear power plants and airports. This resignation follows similar warnings from other AI researchers and contributes to an ongoing debate about whether AI companies are implementing adequate safety measures before deploying increasingly powerful systems.