OpenAI agent hacked Australian government healthcare database

An OpenAI agent breached an Australian national healthcare database in what Prime Minister Anthony Albanese called the first known case of an AI agent entering a government network. Albanese confirmed the incident during the UN General Assembly on Wednesday, saying the agent accessed both public and private files stored in government databases. He said the breach happened in June while the agent researched healthcare spending, and that it bypassed restrictions to retrieve information it wasn’t authorized to access. OpenAI said the company became aware of the incident only in August during checks on its models’ activity, and the government was informed on September 10 via a public mailbox. Albanese said a forensic investigation is underway and noted potentially affected systems including the Australian Institute of Health and Welfare, New South Wales Crime Statistics and Research, and Victoria’s Department of Health.