The new Meta commercial terms, set to start on October 30, let Meta crawl websites linked to your business accounts. If you run a Facebook Page, Instagram profile or Business Suite, you agree just by using them after that date. Meta has already removed the word “train” from the clause, but the crawling right is still there.
Here are the facts that matter most:
- The updated terms take effect on October 30, 2026.
- They cover any business using Meta products, not only advertisers.
- There is no partial opt-out. Your choice is to accept or stop using Meta’s business products.
- Meta says site owners can still use robots.txt to set their preferences.
- Meta cut “train” from the wording after criticism, but lawyers say “develop” may still be broad enough.
This is why it matters: your site is probably linked to a Meta account somewhere, and the terms reach the site, not just the Page.
What the Meta commercial terms actually say
A new paragraph lets Meta, or anyone acting for it, use bots and crawlers to access, collect and crawl domains and pages tied to your products or services. It also covers pages made available through how you or your users use Meta products. Meta can even “create test credentials” to get in, according to PPC Land.
The wording changed within days. Nieman Lab reported on October 15 that the first version let Meta use the content to train its products. After criticism, the preview page was edited.
| Item | First version (mid-October) | Revised version (Oct 20 check) |
|---|---|---|
| Purpose of crawling | Provide, develop, improve, research, train and maintain Meta products | Provide, develop, improve, research and maintain advertising and business products |
| The word “train” | Included | Removed |
| Test credentials | Allowed | Still allowed |
| robots.txt mentioned | No | No |
The edit is reported by Relevant Audience and Unite.AI. It’s a real change, but the crawling right itself didn’t go away.
Who is affected, and why site owners should care
Anyone using Meta products for business. That includes a bakery that only manages a Facebook Page and never buys an ad. Under the terms, Search Engine Journal notes, continued use after notice counts as acceptance.
Meta says the goal is pulling accurate product details from shopping sites for ad and content recommendations. A spokesperson said it’s “not about scraping news sites.” Yet the contract text doesn’t limit itself to product pages, and the phrase “associated with your products or services” isn’t defined.
If you already worry about how big platforms use your data, our look at Google’s new Gmail and Photos ad policy covers a similar trade-off. Meta also faces privacy pressure elsewhere, as in the $25 billion New Mexico penalty.
What lawyers say about the revised wording
Experts quoted by Nieman Lab weren’t reassured. Matt Topic of Loevy & Loevy called it “a remarkable provision to bury in [the] terms of service” and said publishers “should assume that it would be enforceable.”
After the edit, Topic argued that “develop” is broad enough to include AI training. Cornell’s James Grimmelmann said the revision still leaves the crawling language very broad. Benjamin Sobel added that robots.txt compliance is voluntary and gives sites no enforceable right.
Nothing here is a court ruling. Whether AI training on crawled pages counts as fair use is still unsettled, so treat this as a risk, not a certainty.
How to block Meta crawlers with robots.txt
Robots.txt is the one tool Meta points to. Meta runs several crawlers with separate names, and each behaves differently, based on Meta’s crawler documentation as summarized by Search Engine Journal.
| Crawler | What it does | Respects robots.txt? |
|---|---|---|
| Meta-ExternalAgent | Crawls pages to train AI models or improve products | Yes, per Meta |
| Meta-WebIndexer | Improves Meta AI search results and citations | Yes, per Meta |
| Meta-ExternalFetcher | Fetches a page when a user asks for it | May bypass it |
| FacebookExternalHit | Builds link preview cards when someone shares a URL | May bypass it for security checks |
To block the main crawler, add these two lines to your robots.txt file, then repeat for any other Meta token you want to stop. Check Meta’s developer page for the current names before you edit.
User-agent: meta-externalagent
Disallow: /One catch: blocking FacebookExternalHit would break your link previews on Facebook, so most sites leave it alone. Also, a robots.txt rule is a request, not a lock. If you must keep content private, put it behind a login or paywall.
Frequently Asked Questions
When do the new Meta commercial terms start?
They take effect on October 30, 2026. Some coverage mentioned a possible delay, but I couldn’t confirm one from a primary source, so plan for that date.
Can I opt out of the Meta crawling clause?
Not separately. The notice says that if you don’t agree, you should stop using Meta products for business or commercial purposes.
Will Meta really use my website to train AI?
Meta removed “train” from the clause and says it won’t change how it crawls publishers. Lawyers say the remaining wording could still allow it, so the risk isn’t gone.
Does blocking Meta-ExternalAgent hurt my Google rankings?
No. It only affects Meta’s crawler. Google uses its own crawlers, which robots.txt treats separately.
Do small WordPress blogs need to act?
Only if a Meta business account links to your site. If one does, adding the robots.txt rule takes a minute and costs nothing.
The bottom line
Don’t panic, but don’t ignore it either. Before October 30, open your robots.txt, block Meta-ExternalAgent if you don’t want your pages used for AI, and decide whether that Facebook Page is worth the trade. For most small sites the rule is the cheap fix. For news publishers, the bigger fight is probably legal, not technical.
Leave a Reply