Meta has removed language from its commercial terms of use that raised concerns about the company’s ability to collect website content and use it to develop artificial intelligence systems.

The disputed provision suggested that Meta could use content from the domains, webpages and URLs associated with business accounts to provide, develop, improve, research and train its products and services.

The wording prompted concerns among publishers and legal experts about the potential scope of the company’s data collection practices.

Legal experts interviewed by Nieman Lab described the provision as significant because it appeared to give Meta broad permission to collect content from websites connected to businesses using its services.

Critics were particularly concerned that the wording could allow the company to gather publishers’ articles and other copyrighted material for AI training without obtaining separate permission.

The issue emerged as technology companies continue to face legal challenges over the use of copyrighted content to develop artificial intelligence models.

A Meta spokesperson initially maintained that the revised terms did not represent a change in how the company crawls websites across the internet.
The company subsequently removed the disputed provision, according to a report by The Verge.

Meta said it had listened to concerns and clarified that the crawling terms applied to its advertising and business products rather than representing a broader change in its web crawling practices.

The company explained that the original wording had consolidated existing terms relating to its advertising and business services, but acknowledged the need to make their scope clearer.

Meta said the revised language was intended to prevent confusion about the activities covered by those terms.

The clarification followed growing scrutiny of how technology companies obtain online information and whether publishers have sufficient control over the use of their content in AI systems.
The decision highlights the growing tension between artificial intelligence developers and publishers over data ownership, copyright and online content collection.

Although Meta has successfully defended itself against some copyright related claims, legal disputes involving the use of copyrighted material to train AI systems remain ongoing across the industry.

The removal of the disputed language may address some immediate concerns, but questions about transparency, consent and compensation for content creators are likely to remain important as AI development continues.

Source: TechCrunch