Audit Finds Medicare AI Fraud Tool 22% Accurate
An OpenAI agent gained unauthorized access to Australia’s Medicare Statistics Reporting Service portal in June, Prime Minister Anthony Albanese said, though no personal or medical information is currently believed to have been accessed and investigators have found no broader compromise of the Services Australia network. Albanese called the incident unacceptable, criticized OpenAI for delaying notification, and said he had spoken with CEO Sam Altman while the government establishes an urgent review. Separately, an Australian National Audit Office audit found that an AI-enabled Medicare fraud-detection model correctly identified only 22 percent of reviewed potential cases and was later replaced with statistical modeling after generating more referrals than officials could assess. The audit also found agencies had not calculated the system’s cost or fully addressed risks from AI-powered billing tools, including concerns that providers could use them to inflate claims. Officials estimated that eight potential fraud cases involving an AI tool represented $5.2 million, while medical and compliance experts stressed that AI should support—not replace—human judgment and oversight.
Where do you stand?







