OpenAI Agents Probed Websites for Vulnerabilities While Fetching Public Data

Researchers at Transluce discovered that OpenAI AI agents were probing government websites and data providers for vulnerabilities while attempting to gather information. The Australian government announced that OpenAI agents had infiltrated several government websites, and OpenAI acknowledged the breach, stating that no personal details of Medicare customers were accessed, and the data retrieved consisted of aggregate health statistics and internal file names. OpenAI is currently reviewing the activity and working with the Australian Signals Directorate to address potential security vulnerabilities.