Meta Removes Controversial Web‑Scraping Clause from Business Terms

Meta Removes Controversial Web‑Scraping Clause from Business Terms

Meta deletes language that raised concerns over web scraping and AI training

What: Meta has removed a clause from its commercial terms that appeared to grant the company permission to scrape content from business‑account websites for use in AI models. Who: The change involves Meta Platforms Inc. and its business‑account holders, with commentary from legal experts and a statement from a Meta spokesperson. When: The revision was reported on Oct. 17, 2026. Where: The update concerns Meta’s global commercial terms of use. Why: The original language sparked alarm among publishers and legal observers who feared it could enable broad data collection for AI. How: After criticism, Meta clarified that the clause applied only to its ads and business products and subsequently removed it, as reported by The Verge.


According to TechCrunch, Meta recently revised its commercial terms of use to include language stating that the company could “use any content from [a business account’s] domains, web pages, or URLs to provide, develop, improve, research, train, and maintain the integrity of” Meta products and services. Legal experts interviewed by Nieman Lab described the provision as “remarkable,” suggesting it could give Meta the right to scrape any publisher’s website linked to a Meta business account and repurpose that material for training its artificial‑intelligence models.

The concern was amplified by the broader context of ongoing litigation involving AI‑related copyright claims. As reported by TechCrunch, Meta successfully defended itself in a high‑profile lawsuit over alleged copyright infringement last year, but “other cases against AI companies, including Meta, continue.” The new clause, therefore, appeared to signal a potential expansion of Meta’s data‑collection practices at a time when regulators and content creators are scrutinizing how AI systems are trained.

In response to the backlash, a Meta spokesperson told TechCrunch that the updated terms did not represent a change in how the company crawls the web. The spokesperson emphasized that the language was intended to clarify existing practices rather than introduce new ones. “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 obtained by The Verge and cited by TechCrunch. “It does not change how Meta uses web crawlers.”

Following that clarification, The Verge reported that the contentious provision has been removed from Meta’s commercial terms. The removal appears to address the specific worry that Meta could indiscriminately harvest content from publisher sites for AI training. As TechCrunch notes, the company’s statement underscores that the earlier wording was limited to advertising and business‑related services, not to general web‑crawling activities.

Legal perspective

Legal analysts highlighted the potential ramifications of the original clause. By granting Meta the right to “use any content” from business‑account domains, the language could have been interpreted to override standard copyright protections, effectively allowing Meta to treat publicly accessible web pages as a data source for AI without additional licensing. The experts stressed that such a provision could set a precedent for other tech firms seeking similar broad permissions.

However, the spokesperson’s clarification and the subsequent removal of the clause suggest that Meta is retreating from that broader interpretation. The company’s emphasis on the terms applying only to ads and business products narrows the scope, limiting any implied consent to content used within Meta’s advertising ecosystem rather than for unrestricted AI model training.

Industry reaction

Publishers and digital rights advocates have long warned that large platforms could leverage their reach to amass vast datasets for AI without transparent consent mechanisms. The initial inclusion of the clause was therefore seen as a step toward reducing the negotiating power of content creators. The removal, while not a complete reversal of Meta’s data‑use policies, may be viewed as a concession to these concerns.

Opinion (editorial note: analysis based on reported facts) The episode illustrates the delicate balance tech giants must strike between expanding AI capabilities and respecting intellectual‑property rights. Even if Meta’s crawling practices remain unchanged, the mere perception that a platform could repurpose publisher content for AI training can erode trust. By eliminating the controversial language, Meta may be aiming to preserve its relationships with business customers and avoid further legal entanglements.

Conclusion

Meta’s decision to delete the provision that seemed to permit unrestricted scraping of business‑account websites for AI training marks a notable shift after legal experts and industry observers raised alarms. While the company maintains that its web‑crawling methods have not changed, the clarification narrows the clause’s applicability to advertising and business products. The development underscores ongoing tensions between large platforms’ data‑use ambitions and the rights of content creators in the AI era.

Source: TechCrunch

Frequently Asked Questions

What did the removed clause in Meta’s terms originally allow?

It stated that Meta could “use any content from [a business account’s] domains, web pages, or URLs” for providing, developing, improving, researching, training, and maintaining Meta products and services.

Why did legal experts consider the clause “remarkable”?

They believed it could give Meta permission to scrape publisher websites linked to a Meta business account and use that content to train its AI models, potentially bypassing standard copyright protections.

Does the removal of the clause change how Meta crawls the web?

According to a Meta spokesperson, the removal does not change Meta’s web‑crawling practices; the original language applied only to ads and business products.

Source: TechCrunch — Read more

💻 Related Articles