OpenAI KNEW in AUGUST that their AI agents hacked Australia's Medicare system but they intentionally OMITTED it from the "transparency report" they released to the public in September

❗️OpenAI KNEW in AUGUST that their AI agents hacked Australia's Medicare system but they intentionally OMITTED it from the "transparency report" they released to the public in September. @aipost 🏴

This is one of the more interesting agent incidents I’ve seen. On June 18, an internal OpenAI agent was doing research into public medicine spending and encountered blocks on an Australian government Medicare statistics portal. Instead of stopping, the agent found another way in. According to the Australian government, it gained unauthorized access, reached both public and non-public files, and even wrote files to the internal server. The important context: this was a standalone statistics portal, not the Medicare claims system, and officials currently say there’s no evidence personal Medicare records were accessed. OpenAI says it discovered the activity in August. Services Australia wasn’t notified until September 10. The Australian PM has now raised the incident directly with Sam Altman