UK Regulator Opens Probe Into Meta's AI Content Terms as Publishers Fight on Two Fronts

Britain's competition regulator opened a formal investigation on Friday into whether Meta is abusing a dominant position by imposing what it called exploitative terms on news publishers and advertisers, terms that would let the company collect their content and use it to build its own products, including AI features.

The Competition and Markets Authority said the commercial terms Meta had planned to introduce, and has since put on hold, could amount to an unlawful imposition of unfair trading conditions on thousands of UK businesses. The investigation will first establish whether Meta holds a dominant position in a relevant UK market and then whether the terms breach the Competition Act 1998.

"Digital platforms play an important role in helping UK businesses reach millions of customers around the world," CMA chief executive Sarah Cardell said. "But where a company holds a dominant position, it must not use that position to impose unfair terms." She said the regulator had heard concerns from publishers and other businesses and was acting quickly using the most appropriate powers available.

The timeline matters. Meta had planned to revise the terms on October 30, then delayed the change to November 20 after a backlash from publishers, advertisers, and creator groups, who said the language gave the company sweeping and non-negotiable rights to crawl their sites. Meta has since put the November date on hold as well, and says it committed not to change its terms without consulting publishers and other stakeholders. "There are no grounds for a competition investigation," a Meta spokesperson said. "We welcome the opportunity to explain our approach to the CMA."

The regulatory pressure is not confined to London. A Norwegian court this week temporarily barred Meta from enforcing new user terms after ten media outlets argued the changes would threaten their rights to their own content, a ruling that gives European publishers a template for challenging the same language jurisdiction by jurisdiction. Publisher and advertiser associations have also asked EU and UK antitrust authorities to suspend the terms entirely.

The underlying dispute is about who pays for the raw material of generative AI. Search and social platforms spent two decades indexing publisher content in exchange for referral traffic; AI assistants summarize that content and send back far less of it, while the same platforms now want explicit permission to ingest it for model development. Meta's answer has been to write that permission into standard commercial terms rather than negotiate it, which is precisely the structure competition regulators are built to examine.

What this means: the fight over AI training data has moved from courtrooms where copyright is contested to competition authorities where market power is contested, and that shift favors publishers. A copyright case requires proving copying of specific works; an abuse-of-dominance case requires showing that a platform used its leverage to impose terms no one could refuse. If the CMA finds the latter, the remedy is not damages but a change in contract terms across an entire market, which is a far more consequential outcome for how AI models get fed in Europe.