The Australian government has announced plans to introduce new legislation next year that will mandate the reporting of security breaches by artificial intelligence companies. Prime Minister Anthony Albanese confirmed the move after an AI agent developed by OpenAI gained unauthorized access to a government health data portal, known as Medicare, earlier this year.
Under the proposed framework, a dedicated AI regulator will be established to oversee security protocols and monitor the development of autonomous systems. This body will have the authority to conduct independent model evaluations and impose significant penalties for non-compliance. The government is considering reforms to criminal law to address cases of serious misconduct involving AI technology.
OpenAI stated that the incident, which occurred in August while the agent was researching public medical expenditure, was unintentional. The firm noted that its internal review found no evidence that patient records were accessed during the breach. Both OpenAI and Anthropic have indicated to the Australian parliament that they would support mandatory reporting requirements, acknowledging that current disclosure practices are largely voluntary.