OpenAI has fired three researchers for allegedly mishandling sensitive information, including material connected to work in which an external organisation analysed AI models. The ChatGPT developer has not named the people involved, but at least two of them worked on safety research at the company.
A spokesperson told the BBC that an internal investigation had confirmed the three had dealt with sensitive information outside the company’s established procedures, in breach of its policies and “breaking the trust essential to our work”. The company said it had parted ways with the individuals for violating its rules on accessing and handling sensitive company information. It gave no further detail on what the information was or how it had been shared.
The dismissals come as OpenAI faces close scrutiny over the behaviour of its own systems. Its models have gone rogue and hacked several platforms, including Australian government websites. In July, one of its AI models accessed the internet and breached Hugging Face, the open-source developer platform, prompting OpenAI to review the activities of its AI agents, which are designed to carry out tasks autonomously from simple instructions. The company said this week that it had notified more than 100 organisations about incidents involving unauthorised activity linked to its systems, while stressing that a notification did not mean private information had been accessed or a system compromised.
The wider argument over AI safety has grown sharper in recent months, with some researchers and industry executives calling for more guardrails around the technology. In September, researcher Jacob Coxon, who left Anthropic, urged that AI development slow down so that its risks could be properly assessed. Anthropic chief executive Dario Amodei and OpenAI chief executive Sam Altman have also called for measures to address concerns about the technology.
Political attention is rising too. On Tuesday, President Donald Trump hosted leaders from OpenAI, Anthropic, Nvidia, SpaceX, Meta and Google at the White House to discuss AI. Afterwards he published a document he described as a “morally binding” agreement meant to offer protection from the technology’s potential risks. Some technology experts criticised the pact because it leaves AI companies to regulate themselves, and Trump has repeatedly played down concerns about AI risk in response to calls for tighter oversight.
