The UK English edition of Meta’s Commercial Terms preview no longer contains the word “train”. FourWeekMBA read it at 19:05 UTC on 20 October 2026. The sentence that let Meta use content from crawled sites to “provide, develop, improve, research, train and maintain the integrity of” its products, still present when we read the same page at 05:24 UTC on 19 October 2026, is gone.
The UK crawl clause now matches the US English edition’s, which had already dropped “train” when The Verge reported the removal on 17 October. Both editions take effect on 30 October 2026. We compare the old and new wording, set out what did not change, and connect it to our earlier coverage.
Business Pill · CONSENT
A short explainer on consent: you can only agree to what you know. It teaches the general idea only and says nothing about any company in this story.
The key insight: As we read it, Meta has now removed ‘train’ from both English editions of the crawl clause before the terms take effect on 30 October 2026, while keeping the permission to crawl a business’s domains in both.
Update (24 Oct): A US senator asks Meta to withdraw the terms. US Senator Marsha Blackburn, a Tennessee Republican who chairs the Senate Judiciary Subcommittee on Privacy, Technology and the Law, raised concerns on Friday about Meta’s planned commercial terms, saying they could harm businesses and creators, Reuters reported. She said the terms “indicate Meta is attempting to exploit American creators and businesses by granting itself sweeping authority to scrape content from any website associated with Facebook and Instagrambusiness accounts.”
According to Reuters, Blackburn said the earlier removal of an explicit reference to AI training did not go far enough, and that “without clear restrictions on how that information may be used, including whether it may be incorporated into AI systems, Meta’s revisions offer little assurance that creators’ intellectual property will be protected.” She said: “Meta must withdraw the updated Commercial Terms and come to the table to protect the content of creators and publishers across the country.”
Asked about her letter, Meta noted it had delayed the change and said the revised terms apply only to its ads and business products and that it is not changing how it uses web crawlers, Reuters reported. “These updates don’t give us new rights to supersede such preferences and this isn’t about scraping news sites,” the company said, adding that “content owners will still be able to use robots.txt to signal preferences.”
Update (23 Oct): Meta moves the effective date to 20 November.Meta will extend the implementation date of its updated commercial terms to November 20 from October 30 so it can address concerns from publishers and digital advertisers, Reuters reported on Friday. “We are extending the effective date of our updated Commercial Terms and Self-Serve Ad Terms to 20 November to allow more time for us to explain to our partners and stakeholders what’s changing and what’s not,” a Meta spokesperson said in an email, according to Reuters.
The move followed complaints to EU and UK antitrust authorities. On 22 October the advocacy group the Movement for an Open Web, whose members include digital advertisers and publishers, said it had complained to the European Commission and the UK’s Competition and Markets Authority and requested interim measures, Reuters reported, alleging the terms would give Meta an unlimited right to use their content for its AI services. Meta has rejected the claims, Reuters reported; “The crawling terms we consolidated only applied to our ads and business products, and we’ve clarified that in the terms themselves,” a spokesperson said. The CMA said it was “engaged in urgent discussions with Meta”.
Correction note (23 Oct): this piece’s headline and text refer to a 30 October effective date; per Meta’s statement to Reuters, the date is now 20 November. We have not re-read Meta’s terms pages since this change.
On 19 October the UK edition ended its crawling paragraph with this sentence: “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 the same paragraph says Meta may “periodically pull updates from such domains, web pages or URLs in order to provide, develop, improve, research and maintain advertising and business products”, and “maintain the integrity and security of Meta Company Products”. The US edition uses the same words, with American spelling and serial commas.
In our count, the list of permitted uses falls from six verbs to five, and the products it covers change from “the Meta Company Products, services and features provided under these Commercial Terms and any other applicable supplemental terms” to “advertising and business products”, plus the integrity and security of Meta Company Products. Neither edition contains the word “train” anywhere in the text we read.
The crawling paragraph of Meta’s Commercial Terms preview, UK English edition, listed six permitted uses of crawled content including ‘train’ when FourWeekMBA read it on 19 October 2026 (05:24 UTC), and five, without ‘train’, on 20 October 2026 (19:05 UTC), the same as the US English edition. The count of uses is ours, from each clause’s verb list. Both editions take effect on 30 October 2026.
What Did Not Change
Both editions still authorise Meta to use automated software, “such as web scrapers, bots, robots, spiders, crawlers and user agents”, to “access, index, cache, analyse, collect, retrieve or crawl” the domains associated with a business. The effective date is still 30 October 2026.
Meta’s statements since 17 October say the same thing. “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,” the company said in a statement reported by TechCrunch on 17 October. A Meta spokesperson said in a statement reported by Mumbrella on 20 October: “There is no change to how Meta uses web crawlers; any suggestion to the contrary is false.”
Who Pushed Back
The European Publishers Council said the update would allow Meta to “scrape content from news sites”, and said it is writing formally to Executive Vice-Presidents Ribera and Virkkunen. Its executive director, Angela Mills Wade, said: “Maintaining a business account or having a reader share a link should never be treated as a blanket licence to exploit a publisher’s content.”
Mumbrella also quoted Theo Bamber, chief executive of the UK’s News Media Association, who called on the British government to intervene. PPC Land reported on 18 October that the UK edition still carried the sentence.
Update, 21 October: German and UK Publishers Escalate Before Meta’s 30 October Deadline
Germany’s newspaper and digital publishers’ association BDZV and the Medienverband der freien Presse (MVFP) called on the European Commission on 20 October to examine Meta’s change to its commercial terms, due on 30 October 2026, without delay and to prohibit its application until the proceedings are concluded (our translation), according to the BDZV’s press release. In the associations’ words (our translation from German), Meta wants to “automatically capture, copy and store publishers’ complete websites and archives” and evaluate them for its own commercial products, which they call “unscrupulous theft” and, legally, “abusive tying and a breach of the Digital Markets Act”, citing Article 6(12).
The associations added that removing the explicit word “training” from the latest version of the clause “does not solve the problem” (our translation), because other AI uses such as grounding, retrieval-augmented generation, semantic search, summaries and recommendation systems would still be covered. They argued that intervening only after the clause is used would come too late.
In the UK, the campaign group Movement for an Open Web is preparing a legal challenge through the Competition and Markets Authority and the European Commission, ResultSense reported, citing Press Gazette; its director, James Rosewell, said the damage is “imminent and irreversible” once content goes into a model. In the US, Danielle Coffey, president and CEO of the News/Media Alliance, called the terms “outrageous and probably illegal” and said they “cannot be allowed to be implemented on October 30”, Quartz reported, citing Nieman Lab. Meta’s earnings call is on Wednesday, 21 October, Bloomberg reported.
Update, 27 Oct: Norwegian publishers go to court. Members of the Norwegian Media Association, including NRK, TV 2, Schibsted and Amedia, have asked the Oslo District Court to suspend Meta’s new commercial terms pending a judicial review of their legality, Anadolu Agency reported, citing NRK. Our follow-up has the details.
As we read it, the edit narrows the stated purpose rather than the access: the UK clause still lets Meta crawl domains associated with a business, and now ties the uses to advertising and business products and to the integrity and security of Meta Company Products.
As we read it, the gap between the two editions lasted at least from The Verge’s 17 October report to our 05:24 UTC reading on 19 October, and closed by our 19:05 UTC reading on 20 October.
As we read it, for a business on Meta’s platforms the practical text is now one clause, not two, since both English editions use the same words in the crawl paragraph.
Meta, statement reported by TechCrunch, 17 October 2026
“Weโve heard you and want to clarify: the crawling terms we consolidated only applied to our ads and business products, and weโve clarified that in the terms themselves”
Three Implications
ONE CLAUSE IN BOTH EDITIONS The UK English edition read on 20 October uses the same words in the crawl paragraph as the US English edition.
CRAWLING STAYS Both editions still authorise Meta to crawl the domains associated with a business.
30 OCTOBER Both editions take effect on 30 October 2026.
The framework describes the first of its four moats this way: “The Corpus Moat — built by pretraining, eroding as the public web is exhausted and synthetic data closes gaps.”
As we read it, a platform’s commercial terms are one of the places where access to that corpus is written down. The framework was published in July 2026 and does not discuss Meta’s terms.
What Is Not Established
We cannot say exactly when the UK text changed, only that it changed between our two readings. In the sources we read, Meta has not commented on the UK edit specifically. We read the UK and US English editions only, not other language editions. Removing “train” from the clause does not by itself show what Meta does with crawled content, and Meta says its crawling practice is unchanged.
Business Engineer Framework
The Four Intelligence Moats
A Business Engineer map of where AI advantage accumulates, starting with the corpus moat built by pretraining.
By 20 October 2026 the UK English edition of Meta’s Commercial Terms preview, like the US edition, tied the uses of crawled content to “advertising and business products” and the integrity and security of Meta Company Products, and no longer mentioned training. The permission to crawl a business’s domains remains in both, and both take effect on 30 October 2026.
A note on sourcing. This piece is built on FourWeekMBA’s own readings of the UK English edition of Meta’s Commercial Terms preview at 05:24 UTC on 19 October and 19:05 UTC on 20 October 2026, and of the US English edition on 20 October 2026, and on our earlier pieces, which quote The Verge, TechCrunch, PPC Land, the European Publishers Council and Mumbrella. The count of permitted uses in the chart is ours. Nothing here is legal advice.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
