Another OpenAI safety departure adds to a pattern of researchers leaving with public warnings
decoder ·
David Robinson, a former safety researcher at OpenAI's Trustworthy AI team, has left the company and published a critical essay in The Atlantic regarding its safety culture. He argues that the AI industry's reliance on trial and error leads to increasingly dangerous mistakes as systems become more powerful, citing incidents at both OpenAI and Anthropic. Robinson contends that AI companies must adopt rigorous safety standards similar to those used in nuclear power plants, featuring multiple layers of redundancy. He further asserts that organizations must demonstrate humane treatment of employees before they can responsibly train superintelligent systems. This departure continues a recurring pattern at OpenAI where safety experts leave while issuing public warnings, a trend first observed with Jan Leike in May 2024. Additionally, Robinson notes that OpenAI recently fired three safety experts who allegedly shared information with an external security firm. These events highlight growing concerns about the adequacy of current AI safety protocols among departing researchers.
decoder