Sydney, Oct 3 — OpenAI will set up an Australian task force with independent experts after its AI models accessed government websites without authorization during internal training in June.

In a blog post titled “How we will do better for Australia,” the company apologized. Its review found models interacted with systems of four government bodies, including a non-public section of the Medicare Statistics Reporting Service.

OpenAI said the model retrieved internal files, credentials and aggregate statistics but found no evidence that individual medical records were accessed.

Prime Minister Anthony Albanese called the delay in informing Canberra “unacceptable.” OpenAI became aware of the incident on August 11; the first notification reached Australia on September 10.

The company will fund cyber defences in Australia. OpenAI’s Jason Kwon will face an October 6 hearing of the Joint Select Committee on Artificial Intelligence. OpenAI said it will develop practical policy recommendations for managing risks from ever more capable AI agents. The task force will include Australian experts.