Australia announced that an artificial intelligence from the US company "OpenAI" penetrated a government health data portal in June this year, gaining unauthorized access to files, Reuters reported. This may be the first known case of a hacking attack on a government website carried out by artificial intelligence, the agency notes, BTA reported.
This breach is one of the most significant incidents in which AI entities gain access to computer systems outside the US, and comes against the backdrop of a number of recent global breaches carried out by uncontrolled AI. Such examples have raised concerns within government institutions, Reuters reports.
Australian Prime Minister Anthony Albanese said the "OpenAI" agent gained unauthorized access to the medical statistics portal of a government agency responsible for non-confidential health data and statistics, including public health spending.
"The evidence available to date suggests that there has been no larger-scale breach of the network to date. However, this situation is clearly unacceptable", Albanese said at a media briefing in New York, where he was attending the general policy debates of the 81st session of the United Nations General Assembly.
The investigation is ongoing, and Australia has expressed its "deep concern about this incident" to the executive director of "OpenAI" Sam Altman, Albanese said, adding that he was deeply disappointed that the company had been slow to notify the Australian government about the incident.
"It took until September 10th before any notification was sent", the Australian said, adding that the investigation would also look into why government systems did not detect the breach at the outset. He also warned that three other government websites "may have been affected" by the activity of the AI in question.