OpenAI safety systems lead David Robinson resigns after repeated safety concerns

David Robinson, who led OpenAI's safety systems team, has resigned from the company, according to a report from Sina Finance. The departure drew wide attention across Chinese tech media in early October, putting personnel turnover at the AI leader back in the spotlight.

The report ties the resignation to a period in which OpenAI has repeatedly been the subject of safety-related disclosures. After the news circulated, concerns about the stability of the company's safety apparatus grew noticeably across the industry.

According to the report, Robinson had previously written publicly that the current way AI companies are developing their systems is "unacceptable." He was also quoted criticizing the company's culture as having "collapsed," a line picked up by multiple outlets.

On regulation, he argued that large models should be governed at the level of a nuclear power plant, meaning far stronger safety standards and external oversight over how frontier models are built and deployed.

The exit is not an isolated personnel change. Reports indicate OpenAI's safety team has seen continued turnover, and that some employees were dismissed over leaking issues during the same period, again raising questions about team stability.

The resignation of a safety lead matters because it touches a core question for frontier AI companies: whether safety and governance teams can stay independent and stable while the race to ship more capable models accelerates.

What to watch next is whether OpenAI names a successor and how it restructures the safety systems team. The episode may also feed into the broader debate among regulators working on AI rules.