Developing
Whether a court would enforce the clause, whether Meta-ExternalAgent respects robots.txt and how "linked" websites are defined are not stated.
In 60 seconds
- The new Section 1 clause lets Meta use automated software and test credentials to access, index, cache, copy and re-crawl linked sites, and to train its own products on that content.
- Continued use of Meta's commercial products counts as agreement; there is no partial opt-out. It replaces a March 6, 2024 version.
- Meta spokesperson Andy Stone said it is aimed at product information for "products for sale on third-party websites" and "not about scraping news sites".
- Jason Kint of Digital Content Next said it reads as Meta "looking to grant itself the rights to scrape websites across the open web".
Impact Layer
- What changed
- Advertisers accept, by using Meta's products, that Meta may crawl and learn from their linked websites.
- Why it matters
- The clause ties access to a company's own site and content to the use of its ad account, which has copyright and control implications.
- Who is affected
- Advertisers, publishers with ad accounts, legal teams and SEO and GEO specialists.
- Martech stack area
- Advertising, SEO & Search, Content
- Stack impact
- Have legal review the terms before October 30, check whether Meta-ExternalAgent is allowed in your robots.txt and decide what your policy is.
- Business impact
- High PPC Land's reporting on the terms. Whether a court would enforce the clause, whether the terms or crawler honour robots.txt (the terms do not mention it) and how limits on training use would work are not stated.
- Related technologies
- What to watch
- Revisions to the terms and any regulator reaction.
PPC Land reported on October 17, 2026 that Meta’s revised Commercial Terms take effect October 30, 2026.
The clause
- A new Section 1 clause lets Meta use automated software and test credentials to access, index, cache, copy and re-crawl websites linked to advertisers’ products, and to train its own products on that content. It replaces a version dated March 6, 2024.
- Acceptance is passive: continued use of Meta’s commercial products counts as agreement, with no partial opt-out.
- The crawler involved, Meta-ExternalAgent, is described as mixed-use.
- Liability is capped at the greater of $100 or fees paid in the previous twelve months; US businesses face individual arbitration, and non-US disputes can go to courts in California or Ireland.
Positions
Meta spokesperson Andy Stone: the clause is narrower than critics suggest, aimed at product information for “products for sale on third-party websites” and “not about scraping news sites”. Jason Kint, CEO of Digital Content Next: “This reads that Meta is looking to grant itself the rights to scrape websites across the open web.” Attorney Matt Topic: “It’s not clear whether a court would enforce this.” Meta’s share of AI crawling rose from 0.9% to 7.5% between July 2024 and July 2025.
Timing
The terms take effect two days after Meta’s October 28 switch-on of in-app product browsing. A later PPC Land edition noted the training clause remained on Meta’s UK preview page about 30 hours after The Verge reported its removal; the explanations conflict.
Not stated
Whether a court would enforce it, whether the terms or crawler respect robots.txt, how “linked” websites are defined and how training use would be limited.
Our read
Advertisers should read this before October 30. It is a legal change attached to a routine ad account.
Sources
- PPC Land: Meta grants itself the right to crawl advertisers' websites for AI (opens in a new tab)
Source for the clause, statements, figures and gaps.