Click here to sign up for our free daily newsletter
Attendees walk past a Meta logo during the Meta Connect conference, Wednesday, Sept. 23, 2026, in Menlo Park, Calif. (AP Photo/Jeff Chiu)

Canadian news publishers worried about Meta’s new terms of service

Oct 21, 2026 | 2:41 PM

OTTAWA — Canadian news companies say they’re worried about changes to Meta’s terms of service and want the company to clarify how they will be affected.

Meta’s changes to its terms of service, first reported by Neiman Lab, led some to suggest the terms could allow scraping of websites associated with a business account for AI training purposes.

Meta has now revised those terms but the anxiety has not abated. European and British publishers are asking governments and regulators to step in.

Publishers’ association News Media Canada said a statement that “like publisher associations everywhere, we’re very concerned and are seeking clarity.”

The fact Meta blocked news content in Canada following the passage of the Online News Act in 2023 is creating additional uncertainty over how Canadian outlets could be affected by the new terms.

The Canadian Association of Broadcasters said its members have raised concerns and at a minimum, Meta should be transparent about how the new terms will affect Canadian outlets.

Association president Kevin Desjardins said there is an “obvious contradiction” and that Meta has blocked news “arguing that news provides insufficient value to its platforms.”

“Yet it now appears to be seeking broad permissions to access and use content from publishers’ websites for its own commercial purposes,” he said.

Desjardins said “Canadian broadcasters should not be put in a position where their journalism is unavailable to Canadians on Meta’s platforms, while Meta potentially reserves broad rights to exploit it.”

In response to a question about how the terms apply in Canada, a Meta spokesperson said “the crawling terms we consolidated only applied to our ads and business products, and we’ve clarified that in the terms themselves.

“It does not change how Meta uses web crawlers. As always, content owners can use robots.txt to signal their preferences. This is not about scraping news sites, and these updates don’t give us new rights that supersede those preferences.”

Various Canadian news publishers already use robots.txt to block bots from AI companies like OpenAI.

A number of Canadian news outlets, including The Canadian Press, have launched lawsuits against OpenAI and Anthropic over the use of copyrighted content for training AI systems.

This report by The Canadian Press was first published Oct. 21, 2026.

Anja Karadeglija, The Canadian Press