About the hacking of the AI agent of the government website said the Prime Minister of Australia
AI agent of the American company OpenAI hacked the statistical Internet resource of the Australian system of universal health insurance. This was stated by the Prime Minister...

AI agent of the American company OpenAI hacked the statistical Internet resource of the Australian system of universal health insurance. This was stated by the Prime Minister of Australia Anthony Albanese, the correspondent of the agency “Minsk-Novosti”.
Photo by Unsplash
The announcement came at a press conference during the UN General Assembly on September 23, when leading AI companies warned the UN Security Council of global risks and urged governments to work together to maintain control over increasingly powerful AI technologies.
According to Reuters, the site, which was obtained unauthorized access, contains aggregated data on the use of medical services across the country and no personal financial, medical information of millions of Australians. The authorities considered the incident a serious violation. An AI agent looking for health statistics received denials from the system, but did not accept them, and arbitrarily found how to get around the obstacles.
Australia expressed its concern to the management of the company from the United States in connection with the incident. Albanese called the situation unacceptable and noted that it is disappointing that the notification of the incident was received only on September 10 in the form of an email to the agency responsible for public services. At the same time, the country has a national cybersecurity center.
According to media reports, the hack took place in June. The developers, they said, identified the cyber incident later in the summer while checking for instances of inconsistent AI model behavior. Albanese also did not rule out that three other government health websites could have been affected by OpenAI.
The company said it had identified unanticipated AI model activity linked to several Australian government internet services, but found no evidence of access to patients' medical records. In addition, they confirmed their openness and readiness to report such problems and share information.
Australian authorities have launched an investigation into a possible data breach. It also remains to be determined why government systems didn't detect a system security breach in the first place. The challenge is to find out what other measures will prevent such incidents in the future.



