China, Russia “probably” uses AI to try to undermine the next election: CSE-Nation
A new assessment by the Canadian Communications Security Agency said China, Russia and Iran will “probably” use artificial intelligence to try to undermine Canada's next federal election.
Annual report on cyber threats to Canadian democracy released Thursday said cybercriminals associated with these countries are likely to use generative AI tools and methods such as Deepfakes to spread false information, “to aim to sow the division among Canadians and push narratives to benefit foreign interests.”
Canadian politicians and parties may target phishing scams and hacking operations to further sow discord, the report said.
“We assess that China, Russia and Iran will likely use AI-enabled tools to try to intervene in democratic processes before and during the 2025 election,” the report from CSE and its Canadian Centre for Cybersecurity said.
“However, it is unlikely that we will assess that hostile actors will carry out destructive cyber attacks on the electoral infrastructure, such as trying to sanitize telecommunications systems on Election Day (about to emerge or directly armed conflict).
The report also highlights the “higher risk” of Canadian public figures, especially women and LGBTQ2+ people, facing Deepfake porn. It warns that such content can prevent people from running for political office and continue to develop without proper regulations.
“In addition, we may have at least once planted content to intentionally undermine the campaign for office,” the report said.
Get the daily national news
Get news, politics, economics and current events titles delivered to your inbox every day.
The final report on federal public interventions on foreign countries released last month that false information is the biggest threat to Canadian democracy. It says the rise of artificial intelligence is behind the assessment.
The CSE report says that the quality, affordability and accessibility of generative AI technologies enables the use of these tools worldwide.
The agency said that between 2O23 and 2024, more than a quarter of all elections have attempted to use AI's interference and impact, and only once in the past two years has been confirmed.
It noted that during last year's U.S. presidential election, including AI deep strikes by candidates and harassment by U.S. President Donald Trump's campaign for Iran-backed actors, is a well-known example.
The agencies say China and Russia are the greatest backers of opposing global elections in the past two years, while pointing out that they cannot attribute most of these movements to specific players and will continue in the future.
It said that deep strikes and materials generated by foreign participants are often amplified by “extracted or uninformed participants from the target state” and noted that such foreign-generated materials do not gain much appeal on their own.
“CSE is ready to conduct foreign cyber operations to defend our country from hostile threats. Meanwhile, Canadians can help protect democracy by thinking critically about the information they see online,” CSE Chief Caroline Xavier said in a statement to the report.
Meanwhile, the agency said AI is also being used to help foreign countries analyze data collected from target systems faster through large-scale hacking operations.
“National countries, especially Chinese countries, are carrying out large-scale data collection activities, collecting billions of data points about democratic politicians, public figures and global citizens,” the report said.
“Advances in predictive AI have allowed human analysts to quickly query and analyze this data. We evaluate such countries, and therefore may improve people’s understanding of the democratic political environment,” it added, “and they are almost certainly enhancing their ability to engage in targeted influence and espionage.”
A threat announcement issued by the Canadian Cybersecurity Center on Monday warned Chinese-sponsored cyber actors to target Canadian governments, including federal, provincial, municipal and indigenous agencies.
It noted that federal government agencies have been compromised 20 times over the past few years by Chinese cyber threat actors and warned China of “close to situation reconnaissance” activities on the Canadian government system.
“To sum up, Chinese Internet participants have both the number of resources and complex resources that can pursue multiple government goals in Canada at the same time,” the announcement said.
The foreign intervention investigation follows repeated attempts by global news and other media reports about actors such as China who have repeatedly tried to make attempts to speak in Canadian elections and democratic institutions. The final report confirmed that several so-called interference attempts did occur.
Since then, the government has formulated new measures to monitor and remind the public about foreign intervention attempts, including task forces that monitor threatening elections.
Last month, Chrystia Freeland's free-leading campaign warned that a false propaganda campaign sponsored by China spread false news articles about her on whatsapp.
& Copy 2025 Global News, a division of Corus Entertainment Inc.
