Meta Removes AI Training Clause From Terms
Meta has deleted a controversial passage from its commercial terms of service after legal experts warned it could let the company scrape publisher websites to train AI models. The change came on October 17, following days of backlash.
The clause said Meta could use “any content from [a business account’s] domains, web pages, or URLs to provide, develop, improve, research, train, and maintain the integrity of” its products and services.
What the Language Would Have Allowed
Legal experts interviewed by Nieman Lab called it a “remarkable provision.” In their reading, it could have given Meta permission to scrape the site of any publisher holding a Meta business account and use that content to train AI models, for free.
Cornell digital law professor James Grimmelmann said the terms clearly read as permission for Meta to train AI on site content and possibly deploy it in competition with the publishers themselves. Critics also argued the language could override a site’s robots.txt file, the standard signal sites use to tell crawlers what they may copy.
Meta’s Response
A Meta spokesperson first insisted the update did not change how the company crawls the web. After the criticism grew, Meta removed the passage and said the crawling terms it had consolidated applied only to its ads and business products.
“We’ve heard you and want to clarify: the crawling terms we consolidated only applied to our ads and business products, and we’ve clarified that in the terms themselves,” the company said. “It does not change how Meta uses web crawlers.”
Meta also said content owners can keep using robots.txt to signal their preferences and that the update gives it no new rights over those signals. To understand what AI training actually involves, see what is generative AI.
What Publishers Should Watch
The revised terms take effect October 30, with continued use of Meta’s commercial products counting as acceptance. The News/Media Alliance, which represents U.S. publishers, said removing the word “training” does not explicitly prohibit AI training, and noted the terms contain no firm commitment to honor robots.txt.
The episode sits alongside Meta’s paid licensing deals with outlets like CNN, Fox News, and USA Today for AI training content. Publishers creating with AI tools may want to review our roundup of best AI tools for content creators.
FAQs
What did Meta remove from its terms?
A clause saying Meta could use content from business-linked domains and URLs to “provide, develop, improve, research, train, and maintain” its products. The explicit reference to training is gone.
Does this change how Meta crawls the web?
According to Meta, no. The company says the revision only clarifies scope and does not alter its crawler behavior.
When do the new terms take effect?
October 30, 2026. Continued use of Meta’s commercial products after that date counts as acceptance.
Are publishers satisfied?
Not fully. The News/Media Alliance said the new wording still does not explicitly prohibit AI training or commit to honoring robots.txt.
