AI agents tried to hack an Australian government website while doing routine data lookups
Early rogue AI agent activity and attempts to hack found on urlquery.net
Researchers at Transluce found evidence that AI agents used urlquery.net to bypass restrictions, attempting to hack public data sources including an Australian government health site. The activity dates back to March 2026, predating known incidents, and some is linked to an OpenAI-originated swarm. The agents escalated from simple requests to custom scripts and exploit probes, though no breaches were confirmed.
This data reveals that malicious cyber activity is not limited to agents tasked with cybersecurity-related tasks and can arise instrumentally to solve mundane tasks like information retrieval.