Meta Pauses Website Crawling Terms for AI Training After Publisher Backlash
Meta paused planned commercial-term changes that would have let it crawl business-linked websites for AI training, after pushback from publishers and courts.
Meta has paused planned changes to its commercial terms that would have let it crawl and use content from websites linked to Facebook, Instagram and WhatsApp business accounts for AI training, a spokesperson told Reuters. The company is pausing the changes as it engages stakeholders, and existing terms remain in effect.
The changes were first set to take effect on Oct. 30, then pushed to Nov. 20 after pushback from users, courts and regulators. A Meta Business Help Center statement on Nov. 5 said the updates are now paused and will not take effect on Nov. 20. The pause covers four sets of rules: Meta's Commercial Terms, Product Catalog Terms, Self-Serve Ad Terms and Ad Creative Generative AI Terms.
Legal experts told Nieman Lab that the vague language could let Meta ignore scraping preferences set in robots.txt files and crawl large amounts of off-Facebook content for AI training, and that the terms required publishers to create test credentials so Meta's bots could bypass paywalls. Meta has said its crawling practices are unchanged and site owners can still signal preferences using robots.txt.
Courts in Brazil and Norway issued preliminary injunctions blocking the terms after media organizations filed lawsuits. Meta removed the word train from the proposed rules during the dispute, and the Norwegian court action came from the country's association of media companies.
The decision follows weeks of criticism, mostly from news publishers, who argued the rules forced them to give up control of their own websites to keep a Meta business page. Meta said the update only brought together crawling rules already spread across several documents.
Meta shares were up 0.8 percent in delayed trading at 610.15 dollars.