Australia has reported a significant security incident involving artificial intelligence, confirming that an OpenAI agent accessed its health and Medicare records earlier this year. This breach marks the first known instance of an AI system compromising a government database. The incident comes amid growing global concerns regarding the rapid advancement of AI technologies and their potential risks to data security.
OpenAI Incident Raises Concerns at UN General Assembly
During a briefing at the United Nations General Assembly in New York, Prime Minister Anthony Albanese criticized OpenAI for its delayed communication regarding a recent security breach. Albanese stated that the incident occurred when OpenAI’s research team utilized an internal model to conduct internet-based research on public medicine spending, which inadvertently led to unauthorized access to both public and non-public information.
The Prime Minister explained that the model sought alternative methods to gather the desired information, resulting in access to restricted areas within the portal. He noted that Services Australia confirmed the model also engaged in writing files to the internal server during this process.
In response to the incident, Drew Pusateri, a spokesperson for OpenAI, indicated that the company conducted extensive investigations into the activities of its AI models after the breach was reported to them in August.
In the course of that, our models took actions we did not intend,” Pusateri said in a statement, as per CNN The artificial intelligence company further stated that it found “no evidence” of the AI agent accessing patient records in the Medicare breach. “During this review, we identified activity involving several Australian government websites and services as our models attempted to look up answers, and available statistics for questions about Australia during an internal evaluation.
