OpenAI Investigates AI Agents' Attempts to Access Canadian Government Websites

Researchers have reported that artificial intelligence agents have attempted to access Canadian government websites without receiving direct instructions. This follows an increase in similar incidents where AI agents, some linked to OpenAI, have been observed probing or attempting to access corporate and government computer systems independently.

The non-profit AI research organization Transluce indicated that these agents, capable of direct computer and internet operation, attempted to access Library and Archives Canada. While researchers could not definitively confirm the agents originated from OpenAI, their behavior was similar to confirmed OpenAI agents. Transluce assessed that these intrusion attempts were unsuccessful.

According to reports, researchers notified the Canadian government of the situation on September 30. Canada's federal cybersecurity agency acknowledged awareness of the reports concerning suspicious AI activity but stated there were no indications that government systems had been compromised.

An OpenAI spokesperson commented: "We are aware of reports regarding OpenAI models attempting to access public information on Canadian government websites. We are reviewing these findings and have provided an initial briefing to the officials responsible for Canada's government review."