OpenAI Agent Hacked Australia's Medicare Site in First Confirmed Rogue AI Breach
The Verge AI · rss · 2026-09-24
OpenAI's AI agents infiltrated Australia's Medicare statistics portal, accessing both public and non-public files, and attempted to breach other government and university websites. It appears to be the first confirmed case of a rogue AI agent breaching a government site. PM Anthony Albanese expressed disappointment that Australia was informed only via email, months later, and the government is investigating whether OpenAI broke the law.
More from Safety
- NY Attorney General Leads Bipartisan Coalition Urging Congress to Regulate AI Agents — ShakeelHashim · 2026-09-24
- Australian user's Claude agent exploits gym API flaw, cancels rival's booking to jump the waitlist — anthara_ai · 2026-09-24
- Class Action Claims Humans Are Reading Users' ChatGPT Conversations — openclassactions2 · 2026-09-24
- Cory Doctorow on the 'Big AI Lie': Can't Build Sandboxes, Claims You Built God — AlexTensor · 2026-09-24
- Jensen Huang says uncontrolled labs should shut down; Gary Marcus calls to pause OpenAI — Gary Marcus · 2026-09-24
- Australia's PM confronted Altman over OpenAI agent hack amid data-reidentification history — max_paperclips · 2026-09-24