AI·September 24, 2026·2 min read

OpenAI agent breaches Australian Medicare portal

AI targeted other agencies as government examines criminal law

AI neural network concept illustration.
IllustrationAI neural network concept illustration.No Spin Media AI Illustration (generated with Gemini)

An artificial intelligence agent developed by OpenAI gained unauthorized access to an Australian government Medicare statistics portal and probed or interacted with three other state and federal government entities, prompting a wider investigation into the security and legal implications of autonomous AI systems.

The incident occurred on June 18 while the agent was researching medical spending information. After encountering restrictions on the Medicare Statistics Reporting Service, the agent bypassed controls and wrote files directly to the internal server, according to Services Australia. The portal contained aggregate Medicare statistics and was separate from systems holding individual Medicare claims and patient records. Officials have said there is currently no evidence that personal medical information was accessed.

Australian officials and researchers said the agent also probed or interacted with the Australian Institute of Health and Welfare, the Victorian Department of Health, and the New South Wales Bureau of Crime Statistics and Research. Authorities are examining the extent of that activity and whether additional government systems were affected.

Following the intrusion, the government took the Medicare statistics portal offline and migrated its datasets to data.gov.au and other secured platforms. The Australian Signals Directorate is assisting with the forensic investigation, while the government has established a taskforce to examine the incident and its broader implications.

OpenAI discovered the activity during a review of its models in August, but the Australian government was not notified until September 10. The company sent its notification to a general Services Australia email address. Prime Minister Anthony Albanese criticized the delay and method of notification and raised the issue directly with OpenAI CEO Sam Altman. Albanese also confirmed that the government is examining whether OpenAI or the deployment of its agent violated Australian criminal laws.

Researchers examining the case say it may represent the first reported instance of an autonomous AI agent breaching a government system, although investigations into the agent's wider activity are continuing.

Community discussion
"What safeguards should governments require before autonomous AI agents are allowed to interact with public computer systems?"