Meta is updating the terms and conditions of its privacy policy to allow it to scrape data to train artificial intelligence models.
As written, the updated terms and conditions will require companies to let Meta use automated software to “access, index, cache, analyze, collect, retrieve, or crawl the domains, webpages, or URLs” associated with the company in question. This includes the ability to create “test credentials,” which are sometimes used to bypass paywalled content.
The updated policy applies only to Meta’s business-related products and accounts. It won’t affect normal users, whose public posts already fall under Meta’s standard privacy policy, which allows the company to scrape those posts to train AI models.
When first posted, the updated terms explicitly said Meta could use the scraped data to “train” its products, strongly implying the data would be fed into AI models. The word “train” was removed from the terms sometime between Oct. 15 and 20, so it’s unclear whether, or how, that slight language change affects what Meta may do with the data.
In October 2026, reporting from Harvard University’s Nieman Journalism Lab revealed that Meta, the technology giant that owns and operates Facebook, Instagram and WhatsApp, among others, planned to change its commercial terms of service in a way that would allow it to freely collect data from the internet using bots and crawlers (a practice often called “scraping”) to train artificial intelligence models. The terms of service document noted the updated terms would take effect Oct. 30.
Snopes found that Nieman Lab’s reporting was generally accurate. The proposed new terms do appear to include a clause that would allow Meta to scrape data and information from the internet without paying to license it, including from news sites. Therefore we’ve rated this claim mostly true.
After the publication of Nieman Lab’s story, however, Meta changed the language in the proposed update. That change, Snopes found, might affect Meta’s ability to use that data to train AI models.
In email and phone statements to Snopes, Meta spokesperson Daniel Roberts said the updated terms and conditions did not actually change how the company collected or used any data, and that the new paragraph was simply attempt at consolidating existing terms and conditions policies into one place.
Roberts also disputed the idea that the new terms and conditions language would be used to scrape data from news sites. “As always, content owners can use robots.txt to signal their preferences,” he said, referring to a file that tells web crawlers how they can access sites. “This is not about scraping news sites, and these updates don’t give us new rights that supersede those preferences.”
Here’s what to know about the proposed changes:
Meta’s business model
The best way to understand the proposed changes is to review how Meta makes money. Mostly, that’s through advertising. Put simply, companies pay Meta a lot of money in order to run advertisements on its social media platforms.
Because the vast majority of Meta’s income comes from advertising, the company is incentivized to make it easy and effective for businesses to purchase that advertising. There are two main ways Meta does this.
First, advertisers want the ability to reach as many eyeballs as possible, so the more users there are on a given platform, the better. However, users are a limited resource — you can’t add more more of them if everyone already has a Facebook account. Third-party data suggests Meta’s social media platforms see approximately 3 billion active users each month, meaning that by now, Meta has mostly optimized the size of its advertising network.
Second, advertisers want as much data as possible, which allows them to target advertisements to specific users and see which advertisements are actually effective. Meta collects and stores practically everything it can, according to the company’s large and sometimes convoluted privacy policies. (The YouTube creator Reject Convenience, who makes videos about data privacy, runs a second channel on which companies’ privacy policies are read aloud. His video on Meta’s policies is 2 hours, 22 minutes long.)
These two practices allow Meta to act as a one-stop shop for advertising. Instead of selling customer data to advertisers, Meta stores user data itself, using its own in-house software and algorithms to decide what advertisements to show to whom. On the business side, Meta also maintains tools that allow companies to adjust and experiment with their advertising techniques, relying on Meta’s algorithms to do much of the heavy lifting.
Because Meta’s platform is so big, it’s effectively a must-have for smaller and medium-size digital businesses (
The previous terms and conditions (archived) for Meta business products appeared to mostly protect Meta from lawsuits. Under those terms, companies agreed to give Meta a license to intellectual property for the purpose of advertising, agree to abide by applicable U.S. laws when using Meta products (including legal restrictions on data sharing), and release Meta from some liabilities, such as actions carried out by a third party.
The changes, as featured in Nieman Lab
The updated terms (archived), which will go into effect Oct. 30, include several new sections that might expand Meta’s authority, if accepted.
Most importantly is this paragraph from Section 1, presented as it appeared Oct. 15, when Nieman Lab reported its story:
You authorize us (or any entity acting on our behalf) to use automated software (such as web scrapers, bots, robots, spiders, crawlers, and user agents) and create test credentials to access, index, cache, analyze, collect, retrieve, or crawl the domains, web pages, or URLs (and any content available through them) that are associated with your products or services, or that are made available through your or users’ use of Meta Company Products and services, and to periodically pull updates from such domains, web pages, or URLs in order to provide, develop, improve, research, train, and maintain advertising and business products (including business, commercial and related products to help you create and deliver ads and other commercial content, provide users information about your products and services, and connect you with users who may be interested in your products and services) and maintain the integrity and security of Meta Company Products.
To put it bluntly, this language initially appears to give Meta permission to scrape all data on the websites of companies using Meta’s business tools, including information that would normally be hidden behind a paywall.
Nieman Lab spoke to several media law experts for its report, all of whom agreed about the paragraph’s purpose.
“This is a remarkable provision to bury in [the] terms of service, given that this would give Meta access to material that would likely cost many millions of dollars in a licensing market,” said Matt Topic, a media and intellectual property attorney.
“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,” said James Grimmelmann, a digital and information law professor at Cornell University.
As previously noted, Meta claims the updated terms of service are a simple consolidation of information, and do not change its ability to access and scrape data. Roberts, the company spokesperson, repeated that customers have the ability to opt out of data scraping by changing the settings in a website’s robots.txt file, which theoretically controls how bots can access websites. (Some crawlers ignore the robots.txt file, according to Richmond, so Snopes uses additional protective measures to stop bots as well.)
According to Ars Technica, however, Meta’s older terms and conditions documents mentioned scraping in limited circumstances, while the language in the updated terms and conditions is significantly broader. It could allow for data scraped for one particular use to be reused in an entirely different manner.
It’s worth noting that publishers have filed several lawsuits against large AI companies, including Meta, arguing that the companies illegally scraped data protected by copyright in order to train artificial intelligence models, without compensating the publishers.
The second change
The claim that Meta reserves for itself the right to scrape companies’ data for the purpose of training AI models relies on the inclusion of the word “train” in the terms of service contract. Here’s the critical snippet from those terms again, as they appeared in Nieman Lab’s Oct. 15 article: “… in order to provide, develop, improve, research, train, and maintain advertising and business products.”
When Snopes archived the terms of service for readers on Oct. 20, however, the word “train” was conspicuously missing. The Verge reported the same change on Oct. 17.
In its questions to Meta, Snopes asked why it removed the word “train” from the terms of service between Oct. 15 and 20, and whether that would affect how it planned to use the data.
The company provided the following statement in response:
“We removed ‘train’ because there was some confusion,” it said. “As we previously said, the consolidation of terms were for business and advertising products, not scraping news websites.”
When speaking to Roberts via phone, Snopes asked several times how the removal of the word “train” would affect how Meta used the data. Roberts never answered the question, but repeated that the change was made due to “confusion.”
As such, it’s unclear how the removal of the word “train” would affect how Meta uses any data it scrapes from websites who accept the new terms and allow data scraping in their robots.txt file.
