OpenAI safety employee quits, calls for nuclear-level safeguards

David Robinson, formerly leading OpenAI’s safety transparency efforts, wrote in The Atlantic that ChatGPT maker “has thrived by trial and error”