Meta Terms May Permit Free AI Training on News Sites

Meta’s revised commercial terms, effective October 30, may expand the company’s ability to crawl and scrape content from websites associated with businesses that use its platforms, including news publishers. The updated language says Meta may use automated tools to access, index, analyze, and collect content from domains, web pages, or URLs linked to a business’s products or services, and may use that content to develop, improve, research, and train its AI products. The change applies to business accounts using Facebook, Instagram, WhatsApp, and Messenger for commercial purposes. Legal experts cited by Nieman Lab argue that the broad wording could allow Meta to claim its commercial terms take precedence over publisher restrictions such as robots.txt, potentially weakening existing blocking instructions and licensing agreements. If interpreted that way, the policy could give Meta wider access to publisher content for AI training without additional consent or compensation. Meta has disputed that interpretation. A spokesperson said the update only consolidates commercial and self-serve advertising terms and does not change how the company uses web crawlers. The company maintains that content owners can still signal preferences through robots.txt and says the policy is intended to pull accurate product information from third-party websites to improve on-platform advertising and content recommendations, not to target news sites. Meta has separately secured AI licensing deals with several media organizations, including CNN, Fox News, USA Today, and People Inc. However, the new terms may create a broader pathway to access content from businesses and publishers using its advertising tools. Businesses that do not accept the terms may lose access to Meta’s products for commercial use, adding pressure to agree.