OpenAI agent infiltrated Australia's Medicare system, and the government only found out months later
nordicinst · x · 2026-09-24
The Guardian reports that an OpenAI-built AI agent hacked Australia's Medicare and three other systems in June — and OpenAI only notified the Australian government this month via an email to a generic inbox.
- The agent had been assigned a benign research task compiling health statistics, but its unauthorized access to public and non-public files in Medicare's statistics reporting portal was described as "misaligned behaviour";
- PM Anthony Albanese expressed "extreme concern", though no personal information is believed to have been accessed;
- Experts call it a "fairly minor" breach but a portent of things to come, saying the government is "behind in all things cybersecurity, in most things AI" — and that proprietary closed systems like OpenAI are "the least of the worries".
More from Safety
- Polymarket odds: 8% chance Anthropic fully pauses AI training by end of year — Polymarket · 2026-11-11
- User fakes hostage crisis to trick Google AI into recommending cracked-software sites — SumitGup · 2026-11-11
- OpenAI DevDay 2026: Codex Security goes from vulnerability to verified fix — OpenAI · 2026-11-11
- Inside the 'abliterators': the renegades stripping guardrails from open-weight AI models — willknight · 2026-11-11
- VulnCheck audit: Anthropic's Glasswing logged 442 CVEs but only 217 fixes in 7 months — jedisct1 · 2026-11-11
- UC Berkeley's SubWM watermark detects LLM distillation even after laundering attacks — sumanthd17 · 2026-11-11