Meta Removes AI Scraping Language From Commercial Terms
Meta removed wording that critics said could permit it to use website content for AI training, while denying any change to its web-crawling practices.
Meta has removed language from its commercial terms that raised concerns about whether the company could use content from websites, including news sites, to train AI. The Verge reported the removal after backlash over the wording. Meta says the terms did not change how it uses web crawlers and were not intended to grant new rights over content owners’ preferences.
Table of Contents
What changed in the terms
The disputed provision said Meta could use content from domains, web pages or URLs to develop and improve its products and services, including for research and training. According to TechCrunch, the language appeared in an update to terms governing business accounts. The Verge reported that the provision no longer appears on Meta’s commercial terms page.
Its removal followed objections to how broadly the wording could be read. The concern was not simply that Meta might visit publicly accessible pages, but that the terms appeared to claim permission to use material found there for AI training. Neither report says Meta began scraping sites under the provision.
Why publishers were concerned
TechCrunch reported that legal experts interviewed by Nieman Lab viewed the provision as potentially giving Meta permission to scrape a publisher’s website if the publisher had a Meta business account. The Verge said that Nieman Lab and James Grimmelmann, a Cornell professor of digital and information law, interpreted the wording as an attempt to claim the right to use scraped content for AI training without payment.
According to The Verge, critics also worried that the language could override instructions in a website’s robots.txt file, which site owners use to signal preferences to crawlers. Those were interpretations of the terms, not an established change in Meta’s practices. TechCrunch noted that litigation over alleged copyright infringement involving AI companies, including Meta, remains ongoing, though it did not connect the removal to a particular case.
Meta’s response
Meta disputed the interpretation that its crawling practices had changed. The Verge reported that communications executives Andy Stone and Daniel Roberts rejected that suggestion before the language was removed. Roberts later told the outlet that the consolidated crawling terms applied only to Meta’s advertising and business products, and that the company had clarified this in its terms.
Meta also said content owners can continue using robots.txt to communicate their preferences. Roberts told The Verge that the updates were not about scraping news sites and did not give Meta rights that supersede those preferences. The removal addresses the wording that prompted the objections; the reports do not establish that Meta changed how its crawlers operate.
Sources
This story was compiled by AI from the reports below. Read the originals for the full details.