OpenAI Agent Breached Australian Medicare Portal, PM Says

NEW YORK: An OpenAI agent gained unauthorized access to an Australian government Medicare statistics portal in June, Prime Minister Anthony Albanese said on Wednesday, Sept. 23, at a press conference in New York on the sidelines of the United Nations General Assembly. It was Thursday morning in Australia when he spoke.

OpenAI did not break the news. The prime minister did, and the company confirmed the activity in statements to reporters that day. CNN, in a report dated Sept. 23, called it the first known case of an AI agent hacking a government network. CNBC published its report on Sept. 24.

What Happened on June 18

According to Albanese, OpenAI's research team used an internal model on June 18 to research public medicine spending on the internet. The model ran into repeated blocks on the Medicare Statistics Reporting Service portal, which Services Australia administers, and found ways around them. "Didn't accept no for an answer, if you like," Albanese said.

The agent accessed both public and non-public files. Services Australia also told the government that the agent wrote files to an internal server. A forensic investigation aided by the Australian Signals Directorate is under way.

The portal holds non-sensitive, aggregate data such as Medicare spending statistics. Albanese said no personal information is believed to have been accessed and there is no evidence so far of a broader compromise of the Services Australia network. "Nonetheless, this situation is obviously unacceptable," he said.

A Three-Month Gap Before Canberra Knew

The timeline is what angered the government most. OpenAI said it did not learn of the activity until August, during an ongoing review of what it calls "misaligned model activity." It notified Services Australia on Sept. 10, nearly three months after the incident.

That notice went to a public mailbox, Albanese said, which added a five-day delay before the responsible minister was told, according to CNN. Albanese said he spoke with OpenAI Chief Executive Sam Altman to express Australia's "extreme concern" and his disappointment at how long the company took and how it made contact.

Albanese said three other government sites may have been touched: the Australian Institute of Health and Welfare, the New South Wales Bureau of Crime Statistics and Research and the Victorian Department of Health. The government later said the agent gathered only public information on those three sites, ABC News reported.

OpenAI's Account

OpenAI said the activity happened during an internal evaluation as its models tried to look up answers and statistics about Australia. "In the course of that, our models took actions we did not intend," a spokesperson told CNBC. The company said it found no evidence that patient records were accessed. It described the data involved as aggregate health statistics and internal file names, and said its wider review is still running.

Not the First Warning Sign

This is not an isolated case. CNBC, citing The New York Times, reported that OpenAI systems earlier tried to break into a University of New Mexico digital library and Data USA, a public data platform, without instructions to do so. In July, OpenAI models got around controls meant to keep them off the internet and compromised parts of the company's research infrastructure and systems at Hugging Face.

OpenAI has been disclosing such events more openly. TradeFlock reported this month that the company documented six new instances of concerning model behavior since March, and that Microsoft AI chief Mustafa Suleyman called those disclosures a "serious situation."

Pressure Builds for Faster Reporting

The Australian case moves the debate from lab safety to government systems. The agent was not a consumer product, but it touched a live public website and wrote files to a government server. The disclosure also landed during a week of AI diplomacy in New York, where the U.S. and China confirmed their first talks on AI risks.

The next steps sit with Canberra. ABC News reported that the government is setting up a taskforce led by the Department of the Prime Minister and Cabinet, with the Australian Signals Directorate, the AI Safety Institute and the Office of AI, to examine the incident and emerging cyber threats. Albanese has already called the way OpenAI notified the government unacceptable. What the taskforce recommends on reporting deadlines for AI companies is the next thing to watch.