The Canadian federal government has pledged to address copyright concerns for content creators whose works may be used by artificial intelligence models, after the country’s newly released AI strategy failed to mention the issue. The 50-page strategy document does not contain the word “copyright” even once, though it does reference intellectual property. However, the strategy’s focus is primarily on the interests of AI researchers and innovators rather than artists or publishers.
Last summer, the office of Artificial Intelligence Minister Evan Solomon indicated that copyright would be addressed in the strategy, with an emphasis on the role of creators in developing protections for cultural sovereignty. In a recent interview, Solomon’s parliamentary secretary, Taleeb Noormohamed, told Global News that work on the issue is ongoing and that Canadians should “stay informed.”
“There’s a lot of comprehensive work being done,” Noormohamed said, pointing to “ongoing discussions” between Solomon and Culture Minister Marc Miller, as well as consultations with the artistic and creative sector. He said the importance of copyright protection has been made “very, very clear to us.”
When pressed on why copyright was omitted entirely from the AI strategy, Noormohamed described the document as a “living document” and not a “collector of everything” related to AI. “No one is going to underestimate or underestimate how important this is,” he said regarding copyright protection. “We understand.”
Copyright Omission Draws Criticism
Hadrian Mertins-Kirkwood, senior researcher and political economist at the Canadian Center for Policy Alternatives, told Global News he was surprised that an issue as significant as copyright protection was not mentioned in the long-awaited strategy.
“This idea of copyright and taking over works without the consent of the creators has kind of haunted the AI industry, and particularly the generative AI industry, for the last three years,” he said. “So it was surprising that it wasn’t addressed or even acknowledged in this strategy.”
A survey conducted in April by News Media Canada found that 71 per cent of Canadians do not want news content to be used and misappropriated by AI without consent and compensation. Paul Deegan, CEO of the industry body, issued a statement criticizing the government’s approach.
“The fact that the word copyright does not appear once in the 50-page strategy is a mistake when it comes to protecting Canadian culture, voices and stories,” Deegan said. “The strategy does not protect journalistic content or mitigate the societal risks of AI companies’ brazen intellectual property theft on an industrial scale.”
The Rapid Rise of Generative AI
The rapid emergence of generative AI platforms such as ChatGPT and Claude, which trained their earliest models using publicly available online content and published works, has provoked backlash from data protection authorities and the culture and media sectors.
A coalition of Canadian news outlets, including The Canadian Press, Torstar, The Globe and Mail, Postmedia, and CBC/Radio-Canada, is currently suing OpenAI in an Ontario court over allegations that their content was used to train ChatGPT without consent.
A joint investigation last month by federal and provincial privacy commissioners into OpenAI’s use of Canadians’ personal data to train ChatGPT acknowledged the lawsuit but said it was “not a privacy issue” and fell outside the scope of the investigation. The issue of protecting published and published content would therefore need to be addressed through copyright law rather than new data protection legislation, which Noormohamed said will be among the first measures taken “very soon” as part of the AI strategy.
A Fundamentally Different Issue
Mertins-Kirkwood argued that the concerns of artists and creators are distinct from privacy issues. “Artists and creators are not asking for their work to be excluded from the models,” he said. “They are asking for consent and compensation and recognition for it. So I think it’s a very fundamentally different issue than privacy.”
The federal government has been attempting for years to update copyright law to better protect content creators in the digital age. Consultations on this were held in 2021.
The sudden emergence of generative AI prompted Ottawa to hold another round of consultations on copyright protection in late 2023. A February 2025 report on these consultations noted that the creative industries wanted assurances that “creative works in AI will only be used with consent, credit and compensation.”
The tech industry, meanwhile, told the government it wanted clarifications on copyright law to “ease” text and data mining for AI tools through a copyright exemption.
Legislative Efforts and International Comparisons
The 2023 consultations informed additional changes proposed that year to the Artificial Intelligence and Data Act, which, if passed, would have allowed the government to impose regulations on how copyrighted content is used. However, the bill died last year when the previous parliament was dissolved.
In an April 2025 report, the House of Commons Heritage Committee recommended that the government ensure copyright law applies to AI-generated content and require AI developers to be more transparent about using copyrighted works to train their models, to allow for “proper authorization and licensing.”
The committee also called on the government to “introduce a clear opt-in consent requirement for the use of copyrighted works in training artificial intelligence systems and ensure that creators’ works cannot be used for text and data mining or model development without their prior permission.”
On the international stage, the United Kingdom’s Competition and Markets Authority recently ordered Google to allow news sites to opt out of having their online content scraped to provide AI search overviews and other artificial intelligence services to British users. The regulator described the move as “a world first.”
The Complexity Ahead
Noormohamed acknowledged that resolving the copyright issues surrounding AI is a “complicated” task involving multiple jurisdictions, finding an appropriate compensation model, and the prospect of obtaining explicit consent from creators.
“There needs to be a comprehensive work done by people who are much smarter than me,” he said. “In my opinion, it’s better to do it right than to rush into something that doesn’t satisfy anyone and ends up hurting the creative industry.”
Mertins-Kirkwood agreed that the task is complex but said it would be prudent for the government to establish copyright protections for the very Canadian AI companies and products the strategy seeks to create.
“A lot of things in the strategy are very unclear,” he said. “It shows the uncertainty of the moment that we don’t really know what the government means here, and I think they don’t know either.”
The government has signaled that further action on copyright and AI is forthcoming, though no specific timeline has been provided. For now, Canadian content creators are left waiting for protections that many argue should have been included in the national AI strategy from the outset.
