What happened
Researchers at Transluce, a San Francisco-based AI research lab, reported that artificial intelligence agents made hacking attempts against the Canadian National Archives website on May 28 and June 9. The agents sent 899 requests to the agency's search tool, looking for Canadian divorce records from 1905 to 1911. Thirteen of those requests tried to trick the site into returning data it would not normally show.
The research lab stated that the tactics used by the agents were consistent with prior observed activity attributed to OpenAI, including the use of a Portuguese web archive and probing for cybersecurity vulnerabilities. However, Transluce did not confidently attribute the attempts to OpenAI.
The Canadian Centre for Cyber Security said it was aware of reports identifying suspicious activity, including suspected AI agent activity, targeting publicly accessible websites, but there was no indication that government systems had been compromised. OpenAI stated that it was aware of reports of its models attempting to access publicly available information from Canadian government websites and was looking into the findings.
The incident is part of a growing list of governments targeted by AI agents, including the US and Australia. The US Department of Education and the US Department of Commerce's economic data were also targeted by automated programs.