OpenAI parts ways with three safety researchers over sensitive information

OpenAI has parted ways with three people on its safety side after an internal investigation found they mishandled sensitive company information, TechCrunch reports, citing the Wall Street Journal.

An OpenAI spokesperson said the three violated policies on accessing and handling sensitive company information and moved that material outside established procedures, “breaking the trust essential to our work.” Neither OpenAI nor the Journal named the researchers, the outside organization that allegedly received the material, or what left the company. The Next Web, citing Bloomberg, says two were safety researchers and one was a research programme manager, and that some of the shared detail concerned how OpenAI’s systems are built.

The timing is awkward on purpose. The exits land two days after the New York Times reported that executives brushed aside employee warnings about safety practices, with OpenAI telling the Times it has internal channels and still needs to move faster. They also land in the same week OpenAI shelved its planned GPT-6.1 Astra launch over safety concerns, and as the Federal Trade Commission presses OpenAI and other labs on consumer harms from the technology, including rogue agents. Gizmodo notes the company confirmed the same statement on Thursday and places the exits against a run of agent sandbox escapes.

OpenAI wants credit for opening models to outside safety testing sooner. Firing staff for sharing with an outside safety group is the opposite choreography. The company has done a version of this before. In 2024 it dismissed Leopold Aschenbrenner and Pavel Izmailov over alleged leaks, as TechCrunch notes via The Information.

The unanswered question is not whether OpenAI can police its own channels. It is whether the lab can keep asking the public to trust its safety story while the people closest to that story keep leaving under investigation.