Meta removes clause that let it train AI on scraped news sites
Meta deleted the sentence from its Commercial Terms on 17 October 2026, after criticism from the News/Media Alliance, removing the word train from a clause set to take effect 30 October.
Photo: The VergeKey points
Meta removed a clause from its Commercial Terms that critics said allowed it to scrape news sites and train AI on the content, after a day of public criticism.
Meta removed a sentence from its Commercial Terms on 17 October 2026 that had said it could use content from crawled domains to train Meta Company Products. The clause had drawn criticism from the News/Media Alliance and legal scholars, who said it could let Meta scrape news sites. The revised text, in place on Meta's legal website by 17 October 2026, limits the stated purpose of crawling to advertising and business products.
The change matters because the Commercial Terms govern how businesses use Meta's products, and almost every news publisher has at least one Meta business account, according to Nieman Lab. The original wording would have given Meta a contractual right to use scraped content for AI training while business customers had no option to opt out short of ceasing to use Meta's products. The revised clause removes that sentence and the verb train.
What the revised clause now says
The original clause, preserved in an archived capture of the terms page from 16 October 2026, stated that Meta may use any content from crawled domains, web pages, or URLs to provide, develop, improve, research, train, and maintain the integrity of Meta Company Products. A capture from 17 October 2026 shows the sentence gone, and the verb train no longer appears in the provision.
The revised clause still authorises Meta, or any entity acting on its behalf, to use automated software — described as web scrapers, bots, robots, spiders, crawlers, and user agents — and to create test credentials to access, index, cache, analyze, collect, retrieve, or crawl domains, web pages, or URLs associated with a user's products or services. The stated purpose is now to provide, develop, improve, and research advertising and business products, and to maintain the integrity and security of Meta Company Products.
Danielle Coffey, president and CEO of the News/Media Alliance, condemned the original terms in a statement. "Meta's new commercial terms of use are outrageous and probably illegal," she said, adding that withholding access to Meta's audience unless businesses allow scraping is "an appalling, bad-faith policy" and a clear abuse of market power. Coffey said the terms could not be allowed to be implemented on 30 October.
Critics and Meta's response
Meta spokesperson Daniel Roberts said in a statement that the crawling terms consolidated only applied to ads and business products, and that the updates do not give Meta new rights that supersede robots.txt preferences. Andy Stone, Meta's vice president of communications, said the update was designed to retrieve accurate product information for advertising and content recommendations, and rejected suggestions it was intended to enable scraping of news websites.
Eric Goldman, associate dean for research and professor of law at Santa Clara University School of Law, said that the changes raise procedural issues for existing business customers. He said users must be given a bona fide option to reject amended terms, otherwise it is Hobson's choice. Goldman also noted that Meta's language does not expressly say it will respect robots.txt instructions, and that the law does not necessarily compel anyone to respect them.
Meta's Business Help Center post states that updated versions of the Commercial Terms and the Self-Serve Ad Terms will take effect on 30 October 2026. Under the consolidation, the Product Catalog Terms are incorporated into the Commercial Terms, and the Ad Creative Generative AI Terms are incorporated into the Self-Serve Ad Terms. Meta said the changes reduce overlap between its agreements and improve clarity.
What takes effect on 30 October
The Ad Creative Generative AI Terms, in effect since 6 May 2024, already contained a crawling provision allowing Meta to use automated software to retrieve information from domains, webpages, or URLs that advertisers provide. The updated Self-Serve Ad Terms preview, also dated to take effect 30 October 2026, carries the same revised crawling clause with its purpose limited to advertising and business products. There are at least a million businesses using Meta's AI agents in WhatsApp and Messenger, in addition to users of Instagram, Facebook, and Meta's advertising service. The revised Commercial Terms take effect on 30 October 2026, thirteen days after the sentence was removed from the terms page.
Frequently asked questions
What did Meta remove from its Commercial Terms?
Meta removed a sentence that said it may use any content from crawled domains, web pages, or URLs to provide, develop, improve, research, train, and maintain the integrity of Meta Company Products. The revised clause limits the purpose to advertising and business products.
When do the revised Meta terms take effect?
The updated Commercial Terms and Self-Serve Ad Terms take effect on 30 October 2026, according to Meta's Business Help Center post. The revised text was in place on Meta's legal website by 17 October 2026.
Who criticised Meta's original scraping terms?
Danielle Coffey, president and CEO of the News/Media Alliance, called the terms "outrageous and probably illegal" and said they could not be allowed to be implemented on 30 October. Eric Goldman, a law professor at Santa Clara University, raised procedural concerns about existing customers.


