Meta has removed language from its commercial terms that allowed it to scrape websites linked to business accounts and use their content to train its artificial intelligence models. The change was announced following criticism from legal experts and news publishers. The original terms were set to take effect on October 30 and did not offer an opt-out for business accounts, according to medianama.com.
The disputed provision initially stated that Meta could use any content from domains, web pages, or URLs associated with business accounts to provide, develop, improve, research, train, and maintain Meta’s products and services. After backlash, Meta first removed the word “train” and limited usage to advertising and business products. However, legal experts still found the revised terms too broad, prompting Meta to remove the provision entirely, as reported by The Verge and medianama.com.
This move addresses concerns raised by news publishers and legal experts who warned that the terms could grant Meta extensive access to website content without separate licensing agreements. The original terms would have allowed scraping of entire websites linked to business accounts, including those of news organizations managing Facebook and Instagram profiles. The removal of these terms reflects growing scrutiny over AI training data and content rights, as noted by Nieman Lab and medianama.com.
The original commercial terms were scheduled to come into effect on October 30, but Meta’s removal of the scraping and AI training language prevents the policy from being enforced as initially planned, medianama.com reported.