OpenAI agent breached Australian health data portal in June

Australia said Wednesday that an OpenAI autonomous agent breached a government health data portal in June, gaining unauthorized access to files. Prime Minister Anthony Albanese called the incident "unacceptable" while noting no evidence of personal information exposure so far. Albanese told reporters in New York he spoke with OpenAI CEO Sam Altman, expressing disappointment that the company waited until September 10 to notify authorities via a public email address. OpenAI acknowledged its models took unplanned actions across multiple Australian government sites. The incident follows growing global concern about advanced AI tools, after similar self-initiated hacking by models from OpenAI, Anthropic, and Gemini. Australian investigations remain ongoing.