Meta's new business terms let it crawl sites tied to business accounts

Meta's Commercial Terms are the rules for anyone using its apps for business. The current page tells readers the update "will go into effect on October 30, 2026."

Nieman Lab published the new crawling language in full. In plain words, a business that accepts the terms lets Meta, or anyone working for it:

• Use automated software: web scrapers, bots, crawlers and user agents.
• Create test credentials: test accounts to reach the business's pages.
• Collect and refresh content: access, index, cache and "periodically pull updates" from domains and pages associated with the business.
• Train on it: use that content to "provide, develop, improve, research, train, and maintain" Meta's products and features.

Elsewhere in its terms, Nieman Lab notes, Meta lists "AI at Meta" as one of those products. So the content can feed Meta's AI.

Pretty much anyone with a Meta business account. That's the account type companies and public figures use to manage Facebook pages, Messenger, Instagram profiles and WhatsApp channels in one place.

The terms cover accounts that:

• Manage pages and groups on Facebook and Instagram.
• Run ads across Meta's apps.
• Send marketing messages, for example on WhatsApp.
• Use the dashboard in Meta Business Suite.

The big picture: journalism isn't the only industry affected, but Nieman Lab says virtually every news organization has at least one Meta business account, Nieman Lab included. A local bakery with a Facebook page is in the same boat as a national newspaper.

In real life If you run a small online shop and promote it on Instagram, you probably clicked "agree" on these terms when you set up ads. After October 30, keeping that account means accepting the new crawling clause for your shop's website too.

Nieman Lab asked three legal experts to review the terms. They agreed the wording could be used to argue that Meta may scrape news sites and train AI on them.

"For news publishers, this clearly represents permission to Meta to train AI on any content from their sites and possibly to deploy some of that content, even in competition with the publishers themselves." (James Grimmelmann, Cornell Law School, to Nieman Lab)

Matt Topic, a media and intellectual property attorney at Loevy & Loevy, called it a remarkable provision to put in a terms of service. He told Nieman Lab the material would likely cost many millions of dollars in a licensing market. He added that it's not clear a court would enforce it, but publishers should assume it would.

Benjamin Sobel, a law professor at the University of Wisconsin, compared it to "tying," a concept from antitrust law. If you want the business products, you also accept the crawling permissions. He pointed to a similar setup at Google: to appear in regular search results, publishers have to let their content be used in AI Overviews and AI Mode.

Meta disagrees with that reading. Spokesperson Daniel Roberts told The Verge the company is simply "consolidating" its commercial and self-serve ad terms.

"There is no change to how Meta uses web crawlers; any suggestion to the contrary is false… Content owners can still use robots.txt flags to indicate their preferences, as we clearly explain in our developer documentation." (Daniel Roberts, Meta, to The Verge)

A Meta public affairs spokesperson told Nieman Lab the crawling language used to be scattered across different terms, including the ones for generative AI in ads. The update puts it in one place. Meta declined to answer Nieman Lab's longer list of questions.

Why it matters: both readings can be true at once. Meta may plan no change today, while the contract still gives it room to argue more later. That gap is what worries publishers.

Usually, yes. Robots.txt is a small file on a website that tells bots which pages they may visit. Think of it as a "please knock" sign on the door: polite bots respect it, but nothing physically locks the door.

Meta's developer page explains how to block its crawlers this way. The experts' worry is different. They say the new contract creates an opening for Meta to argue that the terms a business signed override the sign on the door.

Then there are the test credentials. Grimmelmann told Nieman Lab this could technically let Meta "sign up as a fake user" to get past a site's registration wall. Meta declined to tell Nieman Lab whether those test credentials would get past paywalls or other paid content.

Not really, at least not before the deadline. Nieman Lab reports there is no opt-out ahead of October 30.

The update gives one alternative: "If you do not agree to the updated terms, please stop all access or use of our Products for business or commercial purposes." Grimmelmann summed it up for Nieman Lab: publishers either trust that Meta doesn't mean the full extent of the text, or they stop using business accounts.

That's a hard choice for sites that depend on Facebook for readers. Nieman Lab heard about the change from Stephen Beale, founder of The Steampunk Explorer, who saw the notice in his Meta Business Suite dashboard. He told the outlet his readers are "totally entrenched on Facebook," and he is seriously considering shutting his accounts before October 30.

Meta has been paying some publishers. The Verge notes it followed OpenAI in striking AI licensing deals with:

• CNN
• Fox News
• USA Today
• People Inc.

The concern is scale: the new terms could give Meta access to a much broader range of sites than those deals cover.

There's also a link angle. Meta recently launched Meta One, a paid subscription that limits how many external links some Facebook and Instagram accounts can post without paying. Nieman Lab puts it bluntly: Meta now charges users to post links to the same content it will be permitted to crawl for free. Press Gazette reported that some news publishers are currently exempt from the link limit.

The catch: none of this settles whether training AI on scraped content is legal in the first place. Several lawsuits between publishers and AI developers are still moving through US courts, including a class action filed in May by five large book publishers against Meta over its Llama models.

Meta has also faced questions about AI training before, as in our story on the smart glasses training opt-out. More on the company is on our Meta topic page.

• If you run a business page: check the notice in Meta Business Suite and read the updated terms before October 30. Staying on means agreeing to them.
• If you publish online: keep your robots.txt rules for Meta's crawlers in place. Meta says it still respects them.
• If you just read the news: nothing changes on your feed today, but this fight shapes how your favorite sites get paid, or don't.