Publishers ask European Commission to stop Meta using news to train AI

Europe’s biggest news publishers have asked the European Commission to intervene against new Meta terms that they say would let the company scrape their journalism to train its AI. The European Publishers Council has written to Commission executive vice-presidents Teresa Ribera and Henna Virkkunen, urging urgent scrutiny before the changes take effect on 30 October, as Decision Marketing reported.

What Meta’s new terms say

Meta’s updated commercial terms apply to any business with a Meta account, and extend to content “made available through your or users’ use” of its products. The EPC argues that wording could let Meta claim permission to scrape news websites, and that a reader sharing a link on Facebook cannot grant Meta rights the reader does not have.

Angela Mills Wade, the EPC’s executive director, said Meta “appears to be using its commercial terms to claim sweeping permission to scrape publishers’ websites”. Her demand is simple: “Opt-in, not opt-out. Licensing, not appropriation.” The group wants the Commission to look at the terms under copyright, contract fairness and competition law. Meta has not commented publicly.

Why it matters

Publishers depend on Meta for traffic but have almost no power to refuse its terms, which is exactly the imbalance the EPC is asking regulators to address. If platforms can turn every shared link into training data, the value of original journalism flows to AI products that compete with the publishers that made it.

It is the latest front in a fight over who gets paid when AI uses other people’s work, and advertisers have a stake in it too: news publishers are among the last large, brand-safe audiences outside the platforms. For more on how platforms are pushing into advertising and AI, see Amazon’s move into ChatGPT ads and our guide to Meta’s ad policies.