OpenAI safety chief resigns over 'unacceptable' development pace

OpenAI's head of safety, David Robinson, has resigned, stating in an open letter in The Atlantic that the company's current direction is "unacceptable." He joins former colleagues in criticizing the pace of development. Robinson argues OpenAI is rushing from one launch to another without the necessary level of caution. He urges AI firms to adopt safety expertise from other fields and wait for research ensuring models make safe decisions without direct oversight. He suggests AI developers should take inspiration from safety and risk management practices used at nuclear power plants or major airports. His departure highlights ongoing internal concerns about the industry's rapid advancement.