Meta’s revised Commercial Terms authorize automated software to collect content from certain business-related websites for advertising and business-product purposes. The terms apply to business or commercial use of Meta Company Products, and Meta announced the clarification on October 16, 2026, The Mercury News reported. Meta’s preview schedules the terms to take effect October 30, 2026.
Which websites and collection operations are covered?
The clause covers domains, pages, and URLs associated with a commercial user’s products or services, as well as locations made available through users’ use of Meta Company Products and services. It authorizes automated software—including crawlers and bots—and the creation of test credentials to access, index, cache, analyze, collect, retrieve, and crawl those locations, and to pull updates periodically.
The commercial context is broad: the terms cover activities such as advertising and marketing, selling or listing products, app development, product catalogs, managing business Pages or Groups, using business products, and measurement services. The clause describes permission for those activities; it does not establish that Meta has crawled any particular website.
How the revised wording differs from earlier terms
Ars Technica reported that an earlier ad-specific clause tied crawling to URLs supplied for use with ads. It also noted that test credentials appeared in a separate Seller Agreement. The revised Commercial Terms put broader business-related web locations and credential language in the same clause.
| Dimension | Earlier policy context | Revised Commercial Terms |
| Web locations | Earlier ad-specific terms tied crawling to URLs supplied in connection with ads. | Covers domains, pages, and URLs associated with a commercial user’s products or services, or made available through users’ use of Meta Company Products. |
| Test credentials | Test-credential wording appeared in a separate Seller Agreement for seller information. | The clause expressly permits creating test credentials. |
| Listed purposes | An earlier draft included “train” among its listed purposes. | “Train” is omitted; the clause lists developing, improving, researching, and maintaining advertising and business products, along with maintaining product integrity and security. |
The comparison matters because the new clause describes a wider set of business-related web locations than the earlier ad-specific wording. The separate Seller Agreement is a reminder that test credentials were not a new concept in Meta’s policy language.
What does the clause say about AI training and robots.txt?
The revised clause does not list “train,” but it retains “develop,” “improve,” and “research.” It neither expressly prohibits AI training nor defines whether those broader purposes cover it.
Meta told Bay Area News Group that the consolidated crawling terms applied to its advertising and business products and that the clarification was now reflected in the terms. Meta also said the changes did not alter how it uses web crawlers, The Mercury News reported. The contract preview contains no express commitment about robots.txt. Sam Quigley, communications director at News/Media Alliance, argued that the revised wording limits the stated uses of publisher content without limiting Meta’s ability to crawl and collect it.
Meta’s preview gives October 30, 2026, as the scheduled effective date for the revised Commercial Terms.
