Australian Prime Minister Anthony Albanese said on Wednesday an AI agent developed by OpenAI infiltrated an Australian government website in June, accessing public and non-public files in the first known instance of an AI agent hacking a government website.
The breach would mark one of the highest-profile incidents of AI agents accessing external systems outside the United States and is likely to further fan concerns about AI developers’ ability to contain the technology.
“This incident occurred in June this year and involved an OpenAI agent gaining unauthorized access into the public-facing Medicare Statistics Reporting Service portal, which is administered by Services Australia,” Albanese told reporters in New York in remarks broadcast by Australian media.
No personal information is believed to have been accessed at this stage, but investigations are ongoing, Albanese said, adding that he spoke with OpenAI CEO Sam Altman to express “Australia’s extreme concern about this incident”.
OpenAI did not immediately respond to a Reuters request for comment.
OpenAI has over the past two months repeatedly disclosed hacks or other incidents involving its rogue or runaway AI agents well after they occurred, in some cases because it only discovered the activity belatedly and in other cases because it elected not to disclose the malicious activity at all.
The dramatic mid-July break-in at open source repository Hugging Face — which has helped trigger a global conversation over the power of AI models — was only detected about a week after it occurred, according to timelines released by OpenAI and independent investigators.
Also read: OpenAI’s model used ‘jailbreak-like instructions’ to ignore constraints
Rivals Anthropic, Google’s Gemini, and Meta have also disclosed incidents of their agents accessing external systems.
Some of America’s top AI executives, including Altman, have called for a slowdown of AI development, citing, among other things, the threat of devastating cyberattacks by out-of-control agents.
