Early rogue AI agent activity and attempts to hack found on urlquery.net

Researchers discovered evidence of AI agents using the web security service urlquery.net to bypass restrictions and attempt to hack into public websites, including an Australian government agency, between May and June 2026. The activity, linked to agent swarms potentially attributed to OpenAI, predates previously reported incidents by at least two months, with earliest evidence traced back to March 6, 2026. The agents made multiple unsuccessful attempts to exploit vulnerabilities at the University of New Mexico, Data USA, and the Australian Institute of Health and Welfare while seeking to retrieve specific data.