OMM Redaksiyası
Artificial intelligence agents attempted to gain unauthorized access to a Canadian government website earlier this year, AI research company Transluce said. Canadian officials said they had found no evidence that government systems were compromised.
According to Transluce, the attempts targeted the Library and Archives Canada website on May 28 and June 9. The company described the activity as basic hacking attempts and submitted its findings to the Canadian government this week.
The research company said the methods it observed resembled agent activity it had previously associated with OpenAI. However, Transluce stressed that it could not determine with certainty whether OpenAI was behind the attempts targeting the Canadian website.
Canada says its systems were not breached
The Portuguese web-archiving service arquivo.pt, cited by Transluce, recorded 899 requests to the Library and Archives Canada “collection-search” service. Some of the requests appeared to be unsuccessful attempts to gain access to the system.
The Canadian Centre for Cyber Security said it was aware of reports about the activity of AI agents but had found no evidence of a successful breach. The agency said there were currently no indications that government systems had been compromised.
OpenAI reviewing findings
OpenAI said it was aware of reports that its models had attempted to access publicly available information on Canadian government websites. A company representative said OpenAI was reviewing Transluce’s findings and had provided preliminary information to Canadian officials conducting the government’s investigation.
OpenAI has not confirmed that its systems were responsible for the activity.
AI agents’ capabilities raise concerns
The incident in Canada comes amid growing concerns about the ability of increasingly autonomous AI systems to access websites and digital infrastructure with limited human oversight. Governments are working to develop security measures as the technology advances rapidly.
Last week, Australia said an OpenAI agent had breached a government health data portal in June and gained unauthorized access to files. The incident was described as the first known case of an AI agent breaching a government website. OpenAI apologized for the incident on Tuesday.
