Meta delays new business terms to November 20: what changed, what the reaction was, and where robots.txt fits in
Topics: AI, Small business

I've already covered what rights Meta gets over business-account websites under its new commercial terms (first article), and how the company dropped the word "train" from the wording after criticism (second article). Now the story has a new twist: the terms themselves, which were supposed to take effect October 30, Meta has delayed until November 20.
Short answer: according to Engadget, a Meta spokesperson told Reuters and Fast Company that the company is pushing back the effective date of its updated commercial terms and self-serve ad terms from October 30 to November 20, 2026, to give itself more time to explain to partners what's changing and what's not. Since the wording was changed, potentially affected entities have filed complaints with antitrust authorities over the updated terms, and the News Media Alliance has demanded that the terms be withdrawn entirely.
What exactly was delayed
The Meta terms draft shown earlier gave the company the right to visit pages tied to business accounts with crawler programs: automated programs that visit a site's pages and copy their content for themselves. The terms list this as the right to access, index, cache, analyze, collect, retrieve, or crawl domains, web pages, and URLs tied to business accounts, and any content available through them. The same version of the terms also said Meta could use that content to provide, develop, improve, research, train, and maintain the integrity of the company's products, services, and features. After criticism from publishers and other businesses, Meta removed specifically the word "train" from the wording (covered in detail in the previous article). Now the company is pushing back the date this already-updated version of the terms (without the word "train") takes effect by three weeks, to November 20.
What the reaction to the terms was
According to Engadget, since the wording changed, organizations the source identifies only as potentially affected have filed complaints with antitrust authorities over the updated terms; the source doesn't specify which organizations, with which authorities, or in which countries. Separately from those complaints, the News Media Alliance, which represents thousands of news organizations in the US and Canada, sent Meta an open letter (not an antitrust complaint) demanding that it withdraw the revised terms of use before October 30. The letter says news publishers would be forced to provide their journalism to support the development of products that could compete with them for readers and advertisers, under duress and without compensation.
Where robots.txt fits in, and why it isn't a guarantee
In its statement, Meta added that it will explain to partners how it uses web crawlers, and will make clear that content owners and publishers can use the robots.txt file. This is an ordinary text file on a site that tells automated programs (crawlers and AI bots) which parts of the site they may visit and which they may not. But according to Engadget, citing an earlier Reuters report, many AI companies' crawlers regularly ignore robots.txt instructions. That means simply setting up this file doesn't guarantee that a site's content won't end up in the data an AI service uses: compliance with robots.txt depends on the good will of the company running the crawler, not on a technical block. The source doesn't say what a site owner should do if a specific crawler doesn't follow the rule, and doesn't name a technical way to force it to: the honest takeaway from this material is that setting up robots.txt communicates a site owner's intent, but doesn't technically block content collection, and the source doesn't go further than that. What you can do right now, without waiting for a reply from Meta: open your own site's address in a browser with /robots.txt added at the end. The file consists of blocks, each starting with a User-agent line naming a bot, or with an asterisk (meaning "for all bots at once"), and the Disallow rules under it apply only to bots in that block. Under the official robots.txt standard (RFC 9309), if there's a separate block with the name of the bot you're interested in, that block applies, not the general asterisk block, even if the general block contains the restriction you need. It's easy to get this wrong when checking it manually for a specific bot name and page, so for that check it's worth using a ready-made tool: for example, the "robots.txt Analysis" tool in Yandex.Webmaster accepts a site address without registration and shows whether a specific page is allowed for the specified bot, rather than just confirming that the file was saved. But even a correct result from that check only means the rule is written properly and asks well-behaved bots not to visit that page, not that it technically blocks access. As for material that can't be handed to crawlers under any circumstances, don't leave it publicly accessible at all: put it behind a real password-protected login and check, without logging in (in incognito mode or logged out), that the page is actually inaccessible, not just hidden from the menu.
Bottom line
Meta pushed back the effective date of its new commercial terms for business from October 30 to November 20, 2026, citing a wish to better explain the changes to partners. After Meta removed the word "train" from the terms, potentially affected organizations filed antitrust complaints, and the News Media Alliance demanded the terms be withdrawn entirely. Meta promises to explain how its crawlers work and points to robots.txt, but according to Reuters as relayed by Engadget, many AI companies' crawlers in practice often ignore that file, so it functions as a signal of intent rather than a technical guarantee.
I run a portfolio of AI agent and automation projects for business: see examples or reach out directly about your own task.
Sources
- Mariella Moon. Meta is delaying its controversial new business terms amid scraping concerns. Engadget, October 24, 2026. https://www.engadget.com/2296124/meta-delaying-controversial-business-terms-scraping-concerns/