The rise in generative AI tools, such as OpenAI’s ChatGPT, has accelerated the proliferation of hyper realistic AI-generated images and video online.Kiichiro Sato/The Associated Press
Nearly nine in 10 Canadians say they are concerned about deceptive online content generated by artificial intelligence, with nearly half encountering misleading or false information daily or several times weekly, according to a new national survey from the Canadian Journalism Foundation.
The survey, conducted on Oct. 21 by the Harris Poll Canada, found that although all age groups are worried about AI-generated misinformation in the news, younger Canadians in particular reported the most frequent exposure to misleading and fabricated content.
The rise in generative AI tools, such as OpenAI’s ChatGPT and Sora, Stable Diffusion and Midjourney, has accelerated the proliferation of hyper realistic AI-generated images and video online, as well as AI-generated text. At the same, with no clear rules around labelling on social media and many other platforms, users are left to figure out on their own what’s real and fake.
The findings from the CJF echo previous surveys about Canadians’ views on misinformation. A Statistics Canada survey from 2023 found that nearly six in 10 Canadians reported they had a very high or extreme level of concern over the presence of misinformation online, while a national survey from August from the non-profit OpenMedia found that 89 per cent of Canadians are worried about AI-generated misinformation, deepfakes, and fake content.
EU won’t ‘lecture’ Canada as it presses ahead with tech and AI regulation, commissioner says
In the survey, respondents said they encountered misleading information in various forms, with “fake news” and clickbait topping the list at 56 per cent and 51 per cent, respectively, followed by deepfakes and altered images (44 per cent), AI-manipulated content (43 per cent) and political disinformation (40 per cent).
Canadians aged 18 to 34 reported the highest recognition of spotting “fake news” at 57 per cent, while respondents aged 61 and older (39 per cent) were least likely to report encountering misleading content.
Natalie Turvey, president and executive director of the Canadian Journalism Foundation, says she found the statistics around deepfakes, altered images and political misinformation the most alarming.
“Several years ago, you wouldn’t have seen such high data points for AI manipulated content,” Ms. Turvey said. “They’re most concerning because they do amplify the risk of misleading news, they erode trust, they create polarization, and they can impair informed decision making.”
Indeed, during the 2025 federal election campaign, an internal briefing note prepared for Canada’s election watchdog and obtained by the University of Ottawa’s Samuelson-Glushko Canadian Internet Policy and Public Interest Clinic, classified the use of artificial intelligence as a “high” risk to violate election rules.
The CJF survey also found that 75 per cent of respondents stated they have some trust in news from traditional Canadian news outlets, with 31 per cent expressing complete trust. By contrast, only 27 per cent expressed trust in social media platforms.
Yet at the same time, Canadians are increasingly getting their news from social media platforms. A year after Meta blocked news articles on Facebook in 2023, three-quarters of Canadians were unaware of the ban, according to a survey from the Media Ecosystem Observatory, a research group at McGill University that analyzes digital threats to democracy.
Fenwick McKelvey, an associate professor in information and communication technology policy at Concordia University, says CJF’s survey results are aligned with Canadians’ general distrust of AI, despite the country being at the forefront of AI research.
“This is something that should be a flag because Canada has purportedly claimed to really invest in AI. So how can we make massive public investments in AI and have widespread public skepticism? One of those two things can’t hold,” Dr. McKelvey said.
He says that one way to address AI skepticism is to hold platforms more accountable when it comes to online advertising, which he says prioritizes content that is the most engaging, regardless if it’s AI-generated or human-generated.
“I think it’s long overdue that we have some accountability on online advertising. What’s fuelling this is the way of monetizing AI slop and the fact that it’s profitable to make junk content online,” Dr. Fenwick said. “We really haven’t had any movement in that.”
In the survey, respondents said that addressing manipulated content online should be a responsibility shared by the federal government (60 per cent), individual Canadians (55 per cent), with social platforms and internet providers also playing a role.
Ms. Turvey says she feels encouraged by the finding that 55 per cent of respondents felt it was their responsibility to learn the tools and skills to combat misinformation.
“I think that’s really positive because it’s an agency mindset that Canadians are showing responsibility, that they know that governments and platforms aren’t going to solve this,” she said.
She’s also heartened by the finding that suggests Gen Z is the age cohort most able to distinguish manipulated and deceptive content. “It’s positive because they are part of our digital future, and it’s really important that they have those discernment skills.”
