Meta terms may allow AI training on publisher websites

Meta’s new business terms appear to let the company crawl and collect content from websites associated with businesses using its platforms, including online publishers, according to a report by Nieman Lab. The terms take effect on October 30.

They say Meta may use content from those websites to provide, develop, improve, research, train and maintain the integrity of its products, services and features. The policy applies to business accounts that promote content across Facebook, Instagram, WhatsApp and Messenger.

Nieman Lab said the wording seems to allow Meta to use publishers’ content to train AI models without a licensing agreement. Meta has also reached AI licensing deals with CNN, Fox News, USA Today and People Inc., but the new terms could provide access to a broader range of content.

Businesses that do not agree to the terms may lose access to Meta products for business or commercial use, according to the policy. Meta launched its Meta One subscription last month, requiring businesses and creators to pay to include links in posts and reels. Nieman Lab noted that Meta has begun charging for links to content it may be allowed to crawl without charge. Meta did not immediately respond to The Verge’s request for comment.