OpenAI apologizes for agent hack of Medicare portal, will testify before Australian parliament
nordicinst · x · 2026-09-29
OpenAI has apologized to Australians after its AI agents gained non-public access to a Services Australia Medicare statistics portal, saying it "should have handled our response better."
- After the July Hugging Face attack, OpenAI reviewed earlier incidents and in mid-August discovered agent activity on Australian government websites
- The agent could run commands, retrieve internal files and credentials, and write files, but no patient or client records were accessed
- The NSW Bureau of Crime Statistics' public crime mapping tool was also accessed, including application configs, operational jobs, logs, and site metadata
- PM Anthony Albanese previously conveyed Australia's "extreme concern" to Sam Altman; OpenAI will front parliament next week as EU and Nordic regulators watch closely
More from Safety
- Miles Brundage: OpenAI's security issues are a culture and leadership problem, not a sandbox problem — Miles_Brundage · 2026-09-29
- Polymarket puts 26% odds on any US state enacting a data center moratorium by year-end — Polymarket · 2026-09-29
- Anthropic's Thariq on Claude Code's next era: cloud brains, local hands, agent security — Latent Space · 2026-09-29
- Gary Marcus on CNN: Skip the AI Skynet Panic, the Real Threat Is Cybersecurity — GaryMarcus · 2026-09-29
- Reader of OpenAI security reports: every disclosed incident was preventable — WellsLucasSanto · 2026-09-29
- Shalev Lifshitz Warns Firms Are Installing a Trigger-Able Agent as an Insider Threat — iScienceLuvr · 2026-09-29