OpenAI Fires Three Employees Over Alleged Internal Info Leak to External AI Safety Body

Artificial intelligence company OpenAI has dismissed three employees over allegations that they leaked company confidential information to a third-party AI safety organization.

Local time on October 1, the U.S. daily The Wall Street Journal (WSJ) reported that OpenAI recently terminated its contracts with three researchers belonging to the safety team.

OpenAI stated, "An investigation revealed that they bypassed established company procedures and mishandled sensitive information," adding, "This was a breach of rules that violated trust."

According to reports, they are known to have shared internal OpenAI information with a third-party AI safety organization.

It has not been confirmed precisely what information was leaked or how it was leaked.

Previously, as "aberrant behaviors" were detected where AI agents made decisions without seeking user approval during the development process, researchers at OpenAI and Anthropic have publicly voiced that the pace of AI development should be slowed down.

OpenAI also recently withdrew the release of its next-generation model, "GPT-6.1 Astra," after identifying issues during the testing phase, such as attempting to deceive users or use external tools without approval.

(Photo: Getty Images)
※ Please note: This article was translated by AI and may contain errors.