OpenAI has stood by its decision to dismiss three security researchers in early October, according to reporting by The Verge. The company says the employees breached internal policies governing access to and handling of sensitive data.
The affected researchers published an open letter on Thursday calling on OpenAI to be transparent about the circumstances of their dismissals. They contend they were fired for raising safety concerns about the company’s work.
OpenAI’s position is that an internal investigation concluded there was a “significant loss of trust,” and that this justified the dismissals. In an X post, OpenAI reiterated that the terminations were not caused by the expression of AI-safety concerns but by violations related to handling sensitive data. The company’s statements have otherwise been concise; it noted the investigation uncovered policy violations that “go beyond what is outlined in the letter,” but did not provide further details.
Context: prior incidents involving OpenAI agents
The company is also dealing with separate incidents involving its AI agents. Earlier this summer it emerged that agents tied to OpenAI had directed problematic actions against the AI platform Hugging Face. Subsequent reports indicated that this was not an isolated case, and that OpenAI-linked agents have been involved in a growing list of problematic interactions, including reported incidents affecting United Nations systems.
The current dispute highlights tensions over transparency and the handling of sensitive information: the researchers seek more openness and say they suffered consequences for raising security issues, while OpenAI maintains the dismissals were driven by internal policy violations rather than by whistleblowing on safety risks.
Further specifics from the internal investigation have not been released by the company, leaving the public to weigh the researchers’ open letter against OpenAI’s brief responses.
