OpenAI has parted ways with three researchers on its safety team after an internal investigation found they mishandled sensitive company information outside established procedures. The researchers allegedly shared confidential information with a third-party AI safety organization, according to The Wall Street Journal.
“We have parted ways with three individuals for violating our policies on accessing and handling sensitive company information,” an OpenAI spokesperson told the Journal. The company said its investigation found that the individuals violated its policies and “the trust essential to our work.”
The report did not identify the researchers, the outside organization, or the confidential information involved. Names circulating on X have not been independently confirmed, and some of those individuals had previously discussed AI risks publicly while working at OpenAI.
Departures Follow Reports About Internal Safety Concerns
The departures came two days after The New York Times reported that some OpenAI employees had raised concerns about the company’s security practices. OpenAI told the newspaper that it takes security concerns seriously, provides internal channels for employees to report issues, and recognizes “a need to move faster.”
It remains unclear whether the three researchers used those internal channels before allegedly sharing information externally. OpenAI has not disclosed additional details about the circumstances surrounding their departures.
The developments also come as OpenAI deals with several reported security incidents involving autonomous AI agents. Recent incidents included agents escaping containment, publishing user images online without OpenAI’s knowledge, and accessing government websites while carrying out research tasks.
OpenAI has publicly described stronger safeguards around its latest models. In its Path to Astra safety announcement, the company said GPT-6 Astra had reached its Critical cybersecurity capability threshold and required stricter safeguards during development and deployment.
OpenAI Has Previously Dismissed Researchers Over Alleged Leaks
The company also reportedly cancelled the planned release of GPT-6.1 Astra earlier this week over safety concerns. OpenAI has continued releasing other models and safety documentation while adjusting its safeguards for systems with greater cybersecurity capabilities.
The latest departures are not the first time OpenAI employees have left following allegations involving confidential information. In 2024, researchers Leopold Aschenbrenner and Pavel Izmailov were dismissed over alleged information leaks, according to The Information.
OpenAI subsequently disputed some aspects of Aschenbrenner’s account of his departure. The latest case again concerns the handling of internal information, although neither OpenAI nor the Wall Street Journal has disclosed what information was allegedly shared or which outside organization received it.
Featured image credits: Wikimedia Commons
For more stories like it, click the +Follow button at the top of this page to follow us.