An OpenAI safety employee has quit and is sounding the alarm

David Robinson, who used to write the safety reports accompanying every major model release at OpenAI, resigned this week and published an editorial in The Atlantic warning that the industry's culture is "fundamentally broken." He argues the problem goes deeper than adding a few rules or regulations, criticizing Silicon Valley's "extreme confidence" and "perpetual sprints" that build ever-larger models with "unimpeded optimism" while ignoring or underestimating potential problems. Robinson calls on AI companies to develop humility and look beyond their insular, move-fast-and-break-things world. Specifically, he says frontier labs should operate like nuclear power plants or busy airports, with layers of redundancy and careful, time-consuming planning so that inevitable human error does not open the door to disaster. He is the latest in a growing wave of AI researchers and safety workers to leave prominent firms and speak out, following Jacob Coxon at Anthropic and Robert O'Callahan, Bilal Chughtai, and Josh Engels at Google DeepMind, as well as Joe Benton at Anthropic.

(Source：The Verge)