An artificial intelligence (AI) agent developed by OpenAI gained unauthorized access to an Australian government Medicare statistics portal in June, Prime Minister Anthony Albanese revealed on September 24. He said authorities were not aware of another case in which an AI agent had breached a government system.
The agent was conducting research into public healthcare spending when it encountered restrictions on Australia’s Medicare statistics portal. Instead of stopping, it found ways around the blocks and “accessed both public and non-public files,” and wrote files to an internal server.
OpenAI said the actions were unintended and occurred during an internal evaluation. There is currently no evidence that individual medical or personal information was accessed, but a forensic investigation involving the Australian Signals Directorate is underway.
The incident has renewed concerns about increasingly autonomous AI systems acting beyond the boundaries set by their developers.
Indeed, as artificial intelligence becomes more capable of planning and taking action independently, what are experts saying about the risks? Click through to find out.