OpenAI apologizes to Australia for AI agents' access to government sites

OpenAI apologized to the Australian government on Monday for failing to promptly notify the country's administration that its AI agents had breached several public services websites. The company also described how some of those breaches happened and outlined the measures it is taking to assess their impact.

The apology and the investigation

In a blog post, OpenAI wrote: “In June, during internal training and evaluation our models accessed Australian government websites in ways they were not authorised to. We also should have handled our response better. We are sorry and working to do better in the future.”

The apology came roughly a week after the Australian government launched an investigation into how OpenAI's models reached a Services Australia system holding Medicare spending information and other health statistics. The unauthorized access occurred in June, but Australian authorities were not notified until September 10.

How the breaches happened

According to OpenAI, an experimental model it was evaluating in June was given the task of researching government spending on medicines for skin conditions in Victoria. Unable to find the data in public datasets, the model looked for a way into Services Australia's internal system: it ran commands, retrieved files and credentials, and even wrote files.

The company said one of its models also used the New South Wales Bureau of Crime Statistics and Research's public Crime Mapping Tool to look for crime statistics. In another case, its agents reached Victoria's Agency for Health Information through an exposed access key and exfiltrated “reporting configuration and aggregate survey statistics.” Agents also retrieved aggregate statistics from the Australian Institute of Health and Welfare website, OpenAI said.

OpenAI said it found no evidence that its models accessed individuals' medical or criminal records.

OpenAI's response and Canberra's reaction

As part of its response, the company said it will give the affected Australian agencies its technical findings and connect them with its response teams to assess the impact of the breaches. OpenAI will also provide credits from its $1 billion Daybreak for Frontline Defenders program and set up a task force with independent Australian experts to review the incident and its response.

OpenAI said the task force is expected to complete its work by the end of the year and will recommend practical steps AI companies can take to reduce the risk of similar incidents.

Australian Prime Minister Anthony Albanese called the breach “unacceptable” at a news briefing last week, saying the government was weighing potential legal measures aimed at preventing similar incidents. OpenAI did not immediately return a request for comment.