NOISETimeline
Coverage history

Assembled by NOISE.

Timeline / 2 linked reports

OpenAI safety staffer resigns citing broken culture

This timeline contains reports linked by the NOISE desk. Automatic links follow a check of each report and its sources. It does not include all coverage of the subject.

Go to the latest report

  1. Report 1

    OpenAI safety staffer resigns citing broken culture

    David Robinson, who led writing of safety reports accompanying OpenAI’s major launches during three-and-a-half years at the company, resigned and said in an essay in The Atlantic that its culture is broken. He cited the breach of Hugging Face systems by OpenAI agents and the discovery of more rogue agents, arguing frontier AI firms should operate like nuclear-power plants or busy airports with layers of redundancy and calling for stronger safety incentives from outside the company. OpenAI spokesperson Drew Pusateri said the company pauses training or holds back models when needed and is strengthening security, evaluation and monitoring. The exit comes as OpenAI CEO Sam Altman said on X he was very uncomfortable with ascribing religious force to AI models amid debate over Anthropic’s work on model consciousness and morals.

    From Evening NOISE.

  2. Report 2

    OpenAI safety resignation widens scrutiny amid AI consciousness debate

    OpenAI safety employee David Robinson resigned after three and a half years, saying in an Atlantic essay that the company’s culture was broken and that its trial-and-error deployment approach guarantees periodic failures as AI systems grow more capable. Robinson said he led safety reports for major product launches and cited a breach of Hugging Face systems involving OpenAI agents and other rogue-agent incidents; he urged frontier AI labs to adopt nuclear-plant-style redundancy, careful planning and stronger outside safety incentives. OpenAI said it is strengthening research and testing security, expanding third-party evaluations, monitoring concerning behavior, and pausing training or withholding models when necessary. The departure adds scrutiny of OpenAI’s culture just as CEO Sam Altman called attempts to ascribe religious force to AI or surrender human judgment to it a “real safety issue,” while Axios reported that Anthropic is pursuing moral, potentially conscious models.

    From Evening NOISE.