TL;DR
- Meta's Commercial Terms preview sets 30 October 2026 as the effective date; the current version dates from 6 March 2024 and has no crawling paragraph.
- The new clause lets Meta or an entity acting for it crawl pages associated with a business, create test credentials, and use the content to train Meta products.
- Business use now explicitly includes managing a Page, marketing messages, product catalogues, Ads Manager and Meta Business Suite, not just ads.
- Nieman Lab's three legal experts (15 Oct) said it could permit AI training on publisher sites; Meta's Andy Stone (16 Oct) called that reading false.
Meta will apply updated Meta Commercial Terms from 30 October 2026, and the new version adds a clause that authorises Meta to crawl websites "associated with" any business that uses its products for a business purpose, and to use that content to "train" Meta products. The text is published on Meta's Commercial Terms preview page. Legal experts quoted by Nieman Lab on 15 October say the wording is broad enough to cover news sites. Meta spokesperson Andy Stone said on X on 16 October that reports framing it that way were "false" and that the update is about pulling accurate product information. This article sets out what the clause says, what changed from the current version, how each side reads it, and what a business in Thailand can check before the date.
What Meta's new crawling clause says, word for word
The preview of the Meta Commercial Terms, effective 30 October 2026, adds a paragraph to the section titled "Licences and rights you provide Meta". Read on Meta's own preview page on 17 October 2026, the first sentence states: "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."
The second sentence covers use of that content: "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."
Four details in that wording matter for anyone reading it closely:
- Who may crawl. The permission covers Meta "or any entity acting on our behalf", so it is not limited to Meta's own staff or systems.
- Test credentials. The clause allows Meta to "create test credentials" to access the pages. The terms do not define what that means in practice.
- Which pages. The scope is pages "associated with your products or services" and pages "made available through your or users' use of Meta Company Products". The second phrase could be read to include links that users, not the business, share on Meta platforms.
- What the content may be used for. The listed purposes are to "provide, develop, improve, research, train and maintain the integrity of" Meta products, services and features provided under the Commercial Terms and supplemental terms.
The preview text does not mention robots.txt anywhere, and it does not limit the crawling permission to product or catalogue pages. Those two points are at the centre of the disagreement described below.
Who the Meta Commercial Terms apply to from 30 October 2026
The Commercial Terms cover far more than advertisers. The preview defines business or commercial purposes as including "using ads or marketing messages, selling or listing products, developing apps, creating or using a product catalogue, managing a Page, managing a group for business purposes, using any business product, such as Ads Manager or Meta Business Suite, or using our measurement services regardless of the entity type." The preview also states that the Meta Terms of Service and the Instagram Terms of Use apply alongside the Commercial Terms for business use.
Acceptance works by continued use. The "Updates" section of the preview says that by continuing any access or use of Meta Company Products for business or commercial purposes after notice of an update, the business agrees to be bound, and: "If you do not agree to the updated terms, you may stop all access or use of our Products for business or commercial purposes." The preview offers no partial opt-out from the crawling paragraph.
In practice, a Thai shop that only runs a Facebook Page, a clinic that sends marketing messages, or a brand that manages a product catalogue in Commerce Manager falls inside the same definition as a company running paid campaigns in Ads Manager.
What changed compared with the current Commercial Terms
The current Meta Commercial Terms, which Meta's live page shows with an effective date of 6 March 2024, contain no crawling paragraph. A check of that page on 17 October 2026 found no reference to scrapers, crawlers or spiders, and no "AI features" section. The 2024 version's "Licences" paragraph covers content a business shares, posts or uploads on Meta products, and the scope paragraph lists "using ads, selling products, developing apps, managing a Page, managing a Group for business purposes or using our measurement services".
Among other changes, the 30 October 2026 version adds three things that were not in the 2024 text: the crawling and training paragraph, a wider list of business uses (marketing messages, listing products, product catalogues, Ads Manager and Meta Business Suite are now named), and a new "AI features" clause.
Search Engine Journal reported on 16 October that Meta's other business terms already contained narrower crawl permissions. According to its report, the Self-Serve Ad Terms dated 22 August 2025 allow crawling of an advertiser's domains when an ad order is placed, with uses that include "research and training"; the Ad Creative Generative AI Terms of 6 May 2024 allow crawling of advertiser-provided URLs for generative AI ad tools; and the Seller Agreement already allows automated access with test credentials to URLs sellers provide. SEJ also reported that the revised Self-Serve Ad Terms carry the same crawl permission from 30 October. Those documents were not reviewed for this article, so those points are attributed to SEJ.
The AI features clause, as written
The 30 October 2026 preview adds a paragraph headed "AI features". It says Meta may use or make available products and features powered by artificial intelligence, machine learning, generative AI or other automated features that help a business understand and use Meta products, or optimise, enhance or create Content. The business acknowledges that "AI features may make mistakes and may not always produce accurate, complete or suitable results."
The clause then places responsibility on the business: it is "solely responsible for and assume[s] any and all risks associated with such AI features and any resulting Content, including evaluating the suitability of any Content for business or commercial purposes." It also states that the business "must not alter or remove any watermark or content-authenticating metadata generated by AI in connection with Content." The preview does not list which AI features this covers.
How legal experts read the crawling clause
Nieman Lab reporters Andrew Deck and Hanaa' Tameez published a report on 15 October 2026 after three legal experts reviewed the terms at the outlet's request. Nieman Lab said it learned of the change from a reader who saw a notice in Meta Business Suite, and that Meta alerted users in early October through a blog post and a Business Suite notice.
The three experts' readings, as reported by Nieman Lab:
- Matt Topic, a media and intellectual property attorney at Loevy & Loevy, said it was not clear a court would enforce the provision but that publishers should assume it would be enforceable.
- James Grimmelmann, a professor of digital and information law at Cornell Law School, said the clause clearly represents permission for Meta to train AI on content from publishers' sites. He also said "test credentials" would technically allow Meta to sign up as a user to get past registration walls.
- Benjamin Sobel, a law professor at the University of Wisconsin, compared tying business product access to the scraping permission to the antitrust concept of "tying".
Nieman Lab also reported that the experts said the language could give Meta an opening to argue that the commercial terms supersede other blocking instructions, and that any AI training would still depend on unsettled legal questions around fair use. These are legal opinions about what the text could permit, not findings about what Meta has done.
Meta's response: crawling practice does not change and robots.txt still applies
Meta gave two public responses. According to Nieman Lab, a Meta public affairs spokesperson said by email that crawling language had previously been scattered across different terms, including terms about using generative AI in ads, and the update consolidates it in one place. The spokesperson said the new terms will not change how Meta crawls publishers and that "content owners can still manage how their content is accessed by using robots.txt." Nieman Lab reported that Meta declined to answer whether test credentials would bypass paywalls.
On 16 October 2026, Andy Stone, who works in communications at Meta, quoted, on X, a Techmeme post summarising the Nieman Lab story. His post, read directly on X, says: "This is false. There's no change to how Meta uses web crawlers." He added that content owners can use robots.txt flags to indicate their preferences, and that the update is designed to make sure Meta can "pull accurate information about products for sale on third-party websites" when making on-platform ad and content recommendations, "not about scraping news sites."
So there are two readings on record. The legal experts read the text for what it would allow. Meta describes what it says it intends to do with it. The published preview itself does not contain the robots.txt commitment or the product-information limit that Meta describes. Whether Meta will or will not train on any particular site is not something the terms, the experts or Meta's statements settle.
Key facts at a glance
The table below sets the main dated facts side by side, each tied to where it comes from.
| Item | What the record shows |
|---|---|
| Effective date | 30 October 2026, per Meta's Commercial Terms preview (Thai version: 30 ตุลาคม 2569) |
| Current version | Effective 6 March 2024; no crawling paragraph and no AI features section |
| New crawling clause | Crawl pages associated with a business; content may be used to "train" Meta products |
| Legal experts (Nieman Lab, 15 Oct) | Wording could permit AI training on publisher content; test credentials could pass registration walls |
| Meta (Andy Stone on X, 16 Oct) | Called reports "false"; said it is about product information and robots.txt flags still work |
Where robots.txt fits: Meta's crawler documentation
Meta's developer documentation on web crawlers, checked on 17 October 2026, lists several separate user agents, each with its own robots.txt token. Meta-ExternalAgent is described as crawling "for use cases such as training foundation AI models or improving products by indexing content directly". Meta-ExternalAds crawls for "improving advertising and other business-related products and services". Meta-WebIndexer is used to improve Meta AI search results, and Meta says allowing it helps it cite and link to content in Meta AI responses.
The same page states two exceptions. Meta-ExternalFetcher "may bypass robots.txt rules" because it fetches links at a user's request, and FacebookExternalHit, the link-preview crawler, "might bypass robots.txt when performing security or integrity checks." Meta also says robots.txt changes can take up to 24 hours to apply because crawlers cache the file.
This is the mechanism Meta points to as the control. What the terms do not say is whether a robots.txt block narrows the permission a business grants by accepting them. That gap is the core of the legal debate. For site owners, how Meta's AI crawlers are treated also affects whether content can be cited inside Meta AI answers, which is part of the wider generative engine optimisation question of which AI systems may read a site.
What this means for Thai marketers
The preview contains no Thailand-specific carve-out, and Meta publishes a Thai-language preview with the same effective date, 30 October 2569 in the Buddhist calendar. 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. The practical points below are checks, not legal advice; a business with specific concerns about content rights should take them to its own lawyer.
1. Map which Meta products the business uses
List every way the business touches Meta for commercial purposes: Pages, Instagram business profiles, ad accounts in Ads Manager, Meta Business Suite, product catalogues, marketing messages, Meta Pixel or Conversions API. Each of these falls inside the preview's definition of business use. Teams running Facebook ads are covered, but so are teams that only manage organic Pages.
2. Decide a position on each Meta crawler in robots.txt
Open the site's robots.txt and check how it treats meta-externalagent, meta-externalads, meta-webindexer and facebookexternalhit. Blocking facebookexternalhit can break link previews on Facebook posts and ads, and blocking catalogue or landing pages from crawlers may affect product information Meta pulls. A blanket decision is rarely right; many sites will want different rules for product pages and for editorial or gated content.
3. Review what sits behind logins and on gated pages
Because the clause mentions test credentials, businesses with member areas, paid content or B2B portals linked from Meta assets should review what those areas contain. Meta has not said whether test credentials bypass paywalls; Nieman Lab reported Meta declined to answer that question.
4. Check how AI-generated ad content is handled
Under the new AI features clause, the business is solely responsible for AI-generated Content and must not remove AI watermarks or content-authenticating metadata. Creative and social ads workflows that edit Meta-generated images or text should be checked against that wording.
5. Get a legal read where content is the product
News publishers are the group the legal experts quoted by Nieman Lab focused on, and other businesses whose website content is itself the product face the same wording. For them, the choice in the terms is binary: accept by continuing business use, or stop business use of Meta products.
What the sources did not say
Several points remain open as of 17 October 2026. Meta has not published a list of which crawler will act under the new clause. Neither Meta's statements nor the preview say whether robots.txt rules override the permission in the terms. Meta has not said which AI products, if any, will be trained on content collected this way. The preview does not define "associated with your products or services". No Thailand-specific guidance has been published. Any claim beyond these points goes past what the record shows.
FAQ: Meta's 30 October 2026 Commercial Terms
When do Meta's new Commercial Terms take effect?
They take effect on 30 October 2026, according to the banner and effective date on Meta's Commercial Terms preview page. The Thai version gives the same date, 30 ตุลาคม 2569.
Do the new terms apply to a business that only has a Facebook Page and does not advertise?
Yes, the preview lists "managing a Page" as a business or commercial purpose. Selling or listing products, marketing messages, product catalogues, Meta Business Suite and measurement services are also listed.
Does the new clause mean Meta will train AI on my website?
The terms permit it, but whether Meta will do so is not established. The clause lists "train" among permitted uses of content from associated sites, and legal experts told Nieman Lab this could cover AI training. Meta's Andy Stone said on 16 October that reports saying so were "false" and that the update concerns product information.
Can I opt out of the crawling clause?
No separate opt-out appears in the preview. The terms say a business that does not agree may stop all access or use of Meta products for business purposes. Meta says robots.txt can still be used to manage crawler access, though the terms text does not mention robots.txt.
Is this live in Thailand?
It applies from 30 October 2026 to business use of Meta products, and the preview contains no Thailand-specific exception. Meta publishes a Thai-language preview with the same effective date.
The terms take effect in under two weeks, so this is a short window to review robots.txt, gated content and the Meta products the business relies on. Relevant Audience can help teams in Thailand audit their Meta setup and crawler rules ahead of 30 October.
