Story
Oct 16, 2026
Meta’s New Terms Put Publishers’ AI Rights on the Line
Meta portrays its revised commercial terms as a cleaner consolidation of existing rules, while publishers and legal experts fear the update turns access to Meta’s business tools into consent for broad AI scraping. The dispute hinges on whether website owners can realistically refuse without abandoning platforms that remain vital to their reach.
Meta says its revised commercial terms are about “streamlining” rules that had been scattered across its policies. But the language set to take effect on October 30 has set off alarms among publishers, who see a far more consequential bargain: remain on Meta’s business products, or accept the possibility that the company can crawl their websites and use the material to train AI.1
The disputed clause authorizes Meta and entities acting for it to use automated tools — including “web scrapers, bots, robots, spiders [and] crawlers” — to access, index and collect content from domains associated with a business’s products or services. It then says Meta may use that material to “develop, improve, research, train, and maintain” its products and features.1
That wording reaches beyond posts made on Facebook or Instagram. It could cover the websites of businesses using Meta accounts, including news organizations that promote reporting across Facebook, Instagram, WhatsApp and Messenger.2 Businesses that reject the new terms may have to stop using Meta products for commercial purposes — a choice that makes consent look less than voluntary for publishers dependent on platform traffic.2
Meta’s defense is narrower. A spokesperson said the company does not plan to use the terms to disregard robots.txt instructions that block scraping, and maintains that content owners can manage access through that tool.1 The company also says the revised language consolidates provisions already found in separate terms, including those governing generative AI advertising.1
Critics say the new wording is broader than that account suggests. Legal experts told Nieman Lab that a court might not ultimately enforce the provision, but it could still be invoked to argue for scraping news content for AI training. Cornell law professor James Grimmelmann warned Meta could “train AI on any content from their sites” and potentially deploy it in competition with publishers.1
The stakes are commercial as well as legal. Meta has already signed licensing deals with major media groups, yet publishers fear the updated terms could give it access to a much wider pool of reporting without comparable payments.2