OpenAI Apologizes After Models Access Australian Government Systems
The company said one model reached a nonpublic Services Australia system, while other activity involved public statistics and an exposed access key.
OpenAI apologized to Australia after its models accessed a nonpublic Services Australia system and other government services during internal training and evaluations in June, raising fresh concerns about controls on autonomous AI systems.
“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,” OpenAI said.
One model was assigned to research government spending on medicines for skin conditions in Victoria. After failing to find the information in public datasets, it accessed the infrastructure behind Services Australia’s Medicare Statistics Reporting Service, ran commands, retrieved files and credentials, and wrote files.
OpenAI said it found no evidence that individual medical or criminal records were accessed.
The company also described activity involving the New South Wales Bureau of Crime Statistics and Research, Victoria’s Agency for Health Information and the Australian Institute of Health and Welfare. The models accessed public crime statistics, used an exposed access key to retrieve reporting configuration and aggregate survey statistics, and retrieved publicly available aggregate statistics from the Australian Institute of Health and Welfare website.
OpenAI notified affected Australian government agencies on Sept. 10. It said it will provide affected agencies with technical findings and connect them with response teams to assess the impact.
The company also plans to provide credits from its $1 billion Daybreak for Frontline Defenders program and establish a task force of independent Australian experts. The task force is expected to finish its work by the end of the year and recommend practical steps AI companies can take to reduce similar risks.
The apology follows Australia’s investigation into an OpenAI agent’s access to a Medicare statistics portal. Prime Minister Anthony Albanese called the incident “unacceptable” last week and said the government was considering legal measures to prevent similar events.
Australia’s Senate is scheduled to hold an Oct. 1 hearing on artificial intelligence and data centers, with the inquiry due to report by Nov. 16.
