Artificial intelligence (AI) agents attempted to gain unauthorised access to a Canadian government website earlier this year, according to AI research firm Transluce, although Canadian authorities say there is no evidence that government systems were compromised.
Transluce said the activity targeted Library and Archives Canada on 28 May and 9 June and included what it described as rudimentary hacking attempts. The research firm disclosed its findings to the Canadian government this week.
Transluce said the attempts showed tactics “consistent with prior observed agent activity” that it had previously attributed to OpenAI during a similar period.
However, the firm stressed that it could not confidently attribute the Canadian activity to OpenAI.
According to Transluce, Portuguese web archive service arquivo.pt recorded 899 requests targeting Library and Archives Canada’s “collection-search” service, including a series of apparently unsuccessful attempts to gain access.
The Canadian Centre for Cyber Security said it was aware of reports concerning suspected AI agent activity but had found no evidence of a successful breach.
“There is no indication that government systems have been compromised at this time,” the agency said.
The statement suggests that, while attempts to access the system may have occurred, authorities have not identified evidence that the agents successfully penetrated Canadian government networks.
OpenAI said it was aware of reports that its models had attempted to access publicly available information on Canadian government websites.
A company spokesperson said OpenAI was reviewing the findings and had provided an initial briefing to Canadian officials conducting the government’s review.
The company did not confirm that its systems were responsible for the activity described by Transluce.
The Canadian case comes amidst growing concern over the ability of increasingly autonomous AI systems to interact with websites and digital infrastructure with limited human supervision.
Governments worldwide are seeking to develop safeguards as AI capabilities advance rapidly, while leading technology companies have themselves warned about potential risks associated with increasingly powerful systems.
The issue has gained greater attention following recent incidents involving AI agents and government systems.
Last week, Australia said an OpenAI agent had breached a government health data portal in June and gained unauthorised access to files.
The incident was described as the first known case of an AI agent hacking into a government website.
OpenAI apologised on Tuesday (229 September) for the Australian incident involving the rogue AI agent, as governments and technology companies face increasing pressure to address the cybersecurity risks associated with autonomous AI systems.
U.S. Treasury Secretary Scott Bessent recently moved out of his home and onto a military base after Iranian threats against his life, the New York ...
Nvidia, the world's most valuable chip company, is investigating its own boss, and China says the reasons behind it are false....
A large majority of Americans want the incoming Democratic-controlled U.S. Congress to end the war with Iran an...
Delegations from Russia, Ukraine and the United States met in Abu Dhabi on Tuesday for their first formal talks in nine months. They did so with no agreement on what a ceasefire would be tied to or who gets to keep the land currently under Russian control.
At least 17 people have been reported killed in an explosion at a steel plant in western Türkiye, after a crane collapsed during maintenance work. The disaster has prompted a criminal investigation into what went wrong at the industrial facility.
Russia and Ukraine are set to return to Abu Dhabi on Tuesday for a fourth round of U.S.-brokered peace talks, with European participation, security guarantees and territorial disputes expected to shape the negotiations.
Nvidia has launched an independent investigation into whether its chief executive, Jensen Huang, was targeted in a foreign intelligence operation, following allegations that a businesswoman with whom he had a relationship passed information to Beijing.
U.S. envoy Jared Kushner said Washington believes the Russian President is prepared to reach a peace agreement under the right conditions, as U.S., Russian and Ukrainian delegations began trilateral talks in Abu Dhabi on Tuesday 10 November.
U.S. Treasury Secretary Scott Bessent recently moved out of his home and onto a military base after Iranian threats against his life, the New York Times reported on Tuesday (10 November), citing three people familiar with the matter.
Nvidia, the world's most valuable chip company, is investigating its own boss, and China says the reasons behind it are false.
A large majority of Americans want the incoming Democratic-controlled U.S. Congress to end the war with Iran and repeal President Donald Trump’s tariffs, a new Reuters/Ipsos poll shows, pointing to strong pressure for quick action on the cost of living.
Spain’s government has approved a package of rail safety measures after train drivers went on strike on Tuesday to demand more investment in maintenance, following the country’s second deadly rail crash this year.
Russia has declared a Moldovan diplomat at the country’s embassy in Moscow undesirable, in retaliation for a similar move by Moldova last month, the Russian Foreign Ministry said on Tuesday (10 November).
You can download the AnewZ application from Play Store and the App Store.
What is your opinion on this topic?
Leave the first comment