TL;DR
- On 20 October 2026 Meta's Commercial Terms preview no longer contained the sentence letting Meta use crawled content to 'provide, develop, improve, research, train and maintain the integrity of' its products.
- The revised clause, effective 30 October 2026, still lets Meta crawl pages associated with a business and create test credentials, now 'in order to provide, develop, improve, research and maintain advertising and business products'.
- Meta spokesperson Daniel Roberts told The Verge on 17 October that the crawling terms only applied to ads and business products; the clause itself still does not mention robots.txt.
- Matt Topic of Loevy & Loevy told Nieman Lab on 17 October that 'develop' is broad enough to include 'train', and two other legal experts it quoted said the revised clause is still broad.
- SEJ reported on 19 October that the UK English Self-serve Ad Terms preview still said 'train'; the same en_GB page showed the revised wording when RA checked on 20 October.
Meta has removed the word "train" from the web-crawling clause in its new Meta Commercial Terms, which take effect on 30 October 2026, but it has kept the right to crawl websites associated with any business that uses its products. When RA re-opened Meta's own Commercial Terms preview page on 20 October 2026, the clause no longer contained the sentence "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. The crawl permission, including the right to "create test credentials", is still there, and the clause now ties the crawling to "advertising and business products" and to "the integrity and security of Meta Company Products". The word "train" does not appear anywhere on the page.
This is a follow-up to RA's post of 17 October 2026, which quoted the original clause as it then appeared on Meta's preview page and set out how legal experts and Meta read it. What is new here is the revised wording, Meta's fuller explanation, the legal experts' reaction to the revision, and a fresh check of Meta's separate Self-serve Ad Terms preview. The background on who the Commercial Terms apply to, the AI features clause and Meta's crawler documentation was covered in the earlier post and is only summarised here.
The clause before and after, word for word
The crawling paragraph sits in the "Licences and rights you provide Meta" section of the Commercial Terms preview. In the version RA quoted from Meta's preview page on 17 October 2026, it had two sentences. The first ended with Meta's right "to periodically pull updates from such domains, web pages or URLs." The second read:
"Meta may use any content from such domains, web pages or URLs to provide, develop, improve, research, train and maintain the integrity of, the Meta Company Products, services and features provided under these Commercial Terms and any other applicable supplemental terms."
On 20 October 2026, Meta's Commercial Terms preview showed a single sentence instead. It keeps the opening permission word for word, then replaces the second sentence with a purpose clause. The revised text, as read on Meta's page, is:
"You authorise 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, analyse, 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 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."
RA checked the English page at facebook.com and at the UK English subdomain, and the Thai page at th-th.facebook.com. All three showed the revised clause, all three gave an effective date of 30 October 2026 (30 ตุลาคม 2569 in the Thai version), and none contained "train" or the Thai equivalent. None of them mentions robots.txt.
What changed and what did not
The table below compares the two versions of the clause point by point, using the text from Meta's preview page.
| Element | Version quoted 17 Oct | Version on 20 Oct |
|---|---|---|
| Crawling and test credentials | Permitted | Permitted, same wording |
| Pages covered | Associated with your products or services, or made available through Meta products | Unchanged |
| Listed uses | Provide, develop, improve, research, train, maintain integrity | Provide, develop, improve, research, maintain |
| Which products | Meta Company Products, services and features under the terms | Advertising and business products, plus integrity and security of Meta products |
| robots.txt | Not mentioned | Not mentioned |
| Effective date | 30 October 2026 | 30 October 2026 |
Three things changed. "Train" is gone. The purposes now attach to "advertising and business products" rather than to Meta Company Products in general, with a parenthetical that describes those as products that help a business create and deliver ads, give users information about its products and connect it with interested users. The integrity purpose became "the integrity and security of Meta Company Products".
Three things did not change. Meta, or any entity acting for it, may still use scrapers and crawlers and "create test credentials" to reach pages associated with a business. The scope still includes pages "made available through your or users' use of Meta Company Products", which could cover links other people share. And the clause still says nothing about robots.txt.
How Meta explained the revision
The revision was reported on 17 October 2026 by Nieman Lab and The Verge. The Verge reported a follow-up statement from Meta spokesperson Daniel Roberts that morning. In it, Meta said: "the crawling terms we consolidated only applied to our ads and business products, and we've clarified that in the terms themselves." The statement also said the update "does not change how Meta uses web crawlers", that content owners can use robots.txt to signal their preferences, and that "This is not about scraping news sites, and these updates don't give us new rights that supersede those preferences."
Nieman Lab, in a report by Andrew Deck and Hanaa' Tameez on 17 October, said a Meta spokesperson explained that the terms were revised because the first consolidation had not made clear that robots.txt preferences are respected. The spokesperson told Nieman Lab that news sites, like other content owners, can still disable the crawler through robots.txt. Nieman Lab also reported that Meta described the change as folding its Ad Creative Generative AI Terms into the general Commercial Terms to make them easier to understand.
Those are Meta's statements about intent. The published clause itself still contains no robots.txt commitment.
Why legal experts say the revised wording is still broad
Nieman Lab went back to the three legal experts it had consulted for its first report on 15 October. According to its 17 October report:
- Matt Topic, a media and intellectual property attorney at Loevy & Loevy, said publishers should remain concerned. He argued that the word "develop" is broad enough to include training, and noted that "business product" does not appear to be a defined term in the terms.
- James Grimmelmann, a professor of digital and information law at Cornell Law School, said the update does not address his concerns. In his view the "develop advertising and business products" language is still very broad, and the edit does nothing to narrow the "associated with your products or services" language that would let Meta crawl entire sites.
- Benjamin Sobel, a law professor at the University of Wisconsin, explained that robots.txt expresses a site's preferences but that compliance is voluntary and gives the site no enforceable legal right.
Nieman Lab also reported that Meta lists examples of its business products as "ads, Meta Business Tools and Meta Business Suite", and that business accounts have no option to opt out of the new terms before 30 October. These are legal opinions and reporting about what the text could permit, not findings about what Meta has done or will do.
Publishers' objections
Several publisher bodies objected to the original wording. Nieman Lab reported that the European Publishers Council said, before the revision, that it would petition EU regulators and the European Commission to scrutinise the terms. The UK's News Media Association issued a statement on 16 October 2026, through its chief executive Theo Bamber, calling for ministers and regulators to intervene. Nieman Lab also reported that the News/Media Alliance in the United States denounced the terms. None of these bodies' reactions to the revised wording were reported in the sources RA reviewed.
The Self-serve Ad Terms: SEJ's 19 October report and a 20 October check
Meta's advertising agreement, the Self-serve Ad Terms, has its own preview with the same 30 October 2026 effective date. Search Engine Journal's Matt G. Southern reported on 19 October 2026 that, as of that day, the UK English preview of the Self-serve Ad Terms still contained the earlier sentence saying Meta "may use any content" from crawled sites to "provide, develop, improve, research, train and maintain the integrity of" its products, while the US English version did not.
RA opened the UK English Self-serve Ad Terms preview, using the same en_GB link SEJ cited, on 20 October 2026. On that day it showed the revised wording, identical to the Commercial Terms clause above, with no occurrence of "train". The Thai Self-serve Ad Terms preview also showed the revised wording. So the discrepancy SEJ reported on 19 October was not present in the versions RA was served on 20 October. RA cannot confirm what other visitors, regions or logged-in accounts are shown, and Meta has already changed these previews once since they were first posted, so the text is worth re-checking before 30 October.
What this means for Thai marketers
The terms contain no Thailand-specific provision, and Meta's Thai-language Commercial Terms preview carries the same revised clause and the same effective date, 30 October 2569. Any Thai business that uses Facebook, Instagram or WhatsApp for a business purpose after that date accepts the updated terms by continuing to use them, and that includes businesses that manage a Page without buying ads. The points below are practical checks, not legal advice.
- Read the clause as it now stands. The AI-training purpose has gone from the text, but the right to crawl pages associated with the business, with test credentials, remains a condition of using Meta's business tools.
- Decide a position on each Meta crawler in robots.txt. Meta points to robots.txt as the control, and its crawler documentation, covered in RA's 17 October post, lists separate user agents for AI training, for ads and business products, and for Meta AI search. Blocking the link-preview crawler can break previews on posts and ads, so changes need care.
- Keep a dated copy of the preview text. Meta revised the clause after publication, and SEJ saw different wording across locales on 19 October. A screenshot or saved copy of the version in force on 30 October is useful if questions come up later.
- Check which teams need to know. Paid social, web, legal and content teams all have a stake, because the clause touches ad accounts, site access and content rights at once.
For teams that run Facebook ads, nothing in the revised clause changes campaign setup, but the robots.txt decision can affect which landing and product pages Meta's crawlers can read. For site owners, how Meta's crawlers are treated also bears on whether content can surface and be cited in Meta AI answers, which is part of the wider generative engine optimisation question, and an SEO audit is a natural point to review crawler rules across all bots.
Frequently asked questions
Did Meta remove AI training from its Commercial Terms?
Yes, the word "train" is no longer in the clause. On 20 October 2026 Meta's Commercial Terms preview listed the purposes as to "provide, develop, improve, research and maintain advertising and business products" and to maintain the integrity and security of Meta products. Matt Topic, one of the legal experts quoted by Nieman Lab, argues that "develop" is broad enough to include training.
Can Meta still crawl my website from 30 October 2026?
Yes. The revised clause still authorises Meta, or anyone acting for it, to crawl pages associated with a business's products or services and to create test credentials to access them.
Does the new clause mention robots.txt?
No. Meta told The Verge and Nieman Lab that content owners can use robots.txt to signal their preferences, but the clause on Meta's preview page does not mention robots.txt.
Is the UK Self-serve Ad Terms preview still different?
Not in the version RA was served on 20 October 2026. SEJ reported on 19 October that the UK English Self-serve Ad Terms preview still had the "train" wording, but the same en_GB page showed the revised wording when RA opened it on 20 October.
Do the terms apply to Thai businesses that do not advertise?
Yes, according to the preview's definition. Business use includes managing a Page, marketing messages, product catalogues and business tools, not only ads, and Meta's Thai-language preview carries the same 30 October 2569 effective date.
Bottom line
Meta has narrowed the stated purpose of its crawling clause and dropped "train" from both the Commercial Terms and, in the versions RA saw on 20 October, the Self-serve Ad Terms. The permission to crawl business-associated sites with test credentials is unchanged, the clause still does not mention robots.txt, and legal experts quoted by Nieman Lab still read it broadly. From 30 October 2026 it applies to every business that keeps using Meta's business products. RA's team can help review crawler settings ahead of that date.
