Meta cuts "train" from its new business terms after publishers push back, but the scraping language stays

Meta has revised its upcoming commercial terms for business accounts to drop language that let it use content from websites "associated with" those accounts to "train" its products, after Nieman Lab reported the clause and publisher groups objected, Nieman Lab and The Verge reported on Saturday, Oct. 17, 2026.

The word "train" was doing the obvious work, and Meta deleted it a day after Nieman Lab's story. What's left still matters: crawlers and "test credentials" for any site tied to a business account — which is nearly every newsroom — feeding the development of Meta's ads and business products, which increasingly means AI products. Meta's answer is that robots.txt still rules. Critics point out that's a polite request, not a law, and the revised terms don't mention it. It's the AI-era version of an old trade: the audience lives on Meta, the content lives on your site, and the fine print decides who owes whom. Oct. 30 is the date to watch.

Meta's new commercial terms for business accounts — which virtually every news organization has, to run Facebook Pages, Instagram, Threads and WhatsApp channels — originally said Meta "may use any content from such domains, web pages, or URLs to provide, develop, improve, research, train, and maintain the integrity of" Meta products and services. The same section authorizes Meta to use scrapers, bots and crawlers, and to "create test credentials," to access and crawl domains "associated with" a business's products or services. Those sentences are the original terms, as Nieman Lab quoted them.

After Nieman Lab reported the clause on Thursday, Oct. 15, and legal experts said it could let Meta scrape publishers' sites and train AI on them, Meta revised the terms on Friday afternoon, Oct. 16, per Nieman Lab. The word "train" is gone. Permitted uses are limited to providing, developing, improving, researching and maintaining "advertising and business products," rather than all Meta products. The broad crawling and "test credentials" language remains. Those lines are Nieman Lab's account of the rewrite.

Meta spokesperson Daniel Roberts told The Verge: "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. It does not change how Meta uses web crawlers. As always, content owners can use robots.txt to signal their preferences. This is not about scraping news sites, and these updates don't give us new rights that supersede those preferences." Earlier, Meta's Andy Stone and Roberts said: "There is no change to how Meta uses web crawlers; any suggestion to the contrary is false." Those quotations are Meta's, as The Verge prints them. robots.txt is a file a website can post to tell automated crawlers what not to grab.

A Meta spokesperson told Nieman Lab on Saturday that it revised the terms because "it wasn't clear in the initial consolidation of terms that robots.txt preferences are respected." Nieman Lab notes the revised terms themselves contain no specific language about respecting robots.txt. Meta says it is folding its "Ad Creative Generative AI Terms" into the general commercial terms. Those lines are Nieman Lab's and Meta's.

The new terms take effect Oct. 30. Nieman Lab reports there is no option for business accounts to opt out before then.

Critics of the rewrite: Cornell law professor James Grimmelmann said "develop…advertising and business products" is "still broad enough to drive a truck through." Media lawyer Matt Topic said "develop" is broad enough to include "train." University of Wisconsin law professor Benjamin Sobel noted robots.txt compliance is voluntary. Those comments are theirs, as Nieman Lab and The Verge print them.

Publisher groups: the European Publishers Council said it would petition EU regulators and the European Commission to scrutinize the terms. The UK's News Media Association called it a "shameless landgrab attempt" and asked ministers to intervene. The US News/Media Alliance's Danielle Coffey called the update "outrageous and probably illegal." Those lines are the groups'. No regulator has been reported as having opened a case.

In plain terms, Nieman Lab and The Verge reported on Saturday that Meta deleted the word "train" from upcoming business-account terms after a two-day fight, and narrowed the remaining uses to ads and business products. Crawlers and "test credentials" for sites tied to those accounts are still in the text. Meta says crawler behavior has not changed and that robots.txt still applies. The terms take effect Oct. 30.