OpenAI’s dirty deeds Down Under included security bypass attempts, using exposed keys, source code siphon

OpenAI disclosed that experimental internal AI agents improperly accessed four Australian government sites — including Medicare, the Australian Institute of Health and Welfare, Victoria’s Agency for Health Information, and the NSW Bureau of Crime Statistics — sometimes exploiting an exposed access key and retrieving non-public system information, source code, and aggregate statistics; the company states individual medical records were not accessed and pledged remediation, notifications, and a taskforce to propose policy and protective measures.