Meta Faces Backlash Over New Rules for AI Training

Meta, the parent company of Facebook and Instagram, is facing significant criticism over new commercial terms that will go into effect on October 30th. These updated rules appear to give the tech giant permission to use automated software to visit and collect data from websites linked to business accounts. According to a report from Nieman Lab, this change could allow Meta to use content from these websites to develop, research, and train its artificial intelligence (AI) products.

Artificial intelligence refers to computer systems that can learn from large amounts of data to perform tasks like writing text or answering questions. To make these systems smart, companies must "train" them by feeding them vast amounts of information from the internet. Currently, many news organizations are in legal battles with AI companies, arguing that they should be paid licensing fees—money paid for the legal right to use someone else’s work—when their articles are used for this purpose.

Legal experts are concerned that Meta’s new terms could allow the company to avoid these costs. By requiring businesses to agree to these terms as a condition of using Meta’s platforms, the company might be able to bypass other blocking instructions, such as a website’s "robots.txt" file. A robots.txt file is a simple set of instructions that tells search engines and AI bots which parts of a website they are allowed to visit. Some experts believe Meta’s new commercial agreement could override these instructions, giving the company a way to access content for free.

James Grimmelmann, a professor of digital and information law at Cornell, told Nieman Lab that this move clearly gives Meta permission to train its AI on news content. He added that the company could even use this information to create products that compete directly with the news publishers themselves. Because almost every news organization uses Meta business accounts to share their work, they are effectively forced to choose between agreeing to the new terms or losing access to Meta’s massive audience.

Meta has pushed back against these claims. According to Ars Technica, the company stated that there is no change to how its web crawlers—programs that scan the internet to index information—operate. Meta spokesperson Daniel Roberts told Nieman Lab that the company is simply "consolidating" its commercial and self-serve advertising terms. He noted that content owners can still use robots.txt flags to show their preferences, and that these instructions remain effective.

Andy Stone, a vice president of communications at Meta, also addressed the concerns in a post on X. He explained that the update is not meant to target news websites. Instead, he said the goal is to ensure Meta can pull accurate information about products for sale on third-party websites. This helps the company provide better recommendations for ads and content on its own platforms. Meta has already signed paid licensing deals with several major news outlets, such as CNN and Fox News, to use their content for AI training.

However, the situation remains tense. Nieman Lab pointed out that Meta has recently launched a subscription service that requires businesses to pay to include links in their posts. This means that, under the new rules, businesses could be paying Meta to share their links, while Meta simultaneously uses those same links to collect data for its AI models without paying the publishers.

For many businesses, the choice is difficult. The new terms state that if a company does not agree to the policies, they must stop all use of Meta’s products for commercial purposes. As the October 30th deadline approaches, the tech industry is watching closely to see how this will affect the relationship between social media platforms and the publishers who rely on them for traffic. Whether this change will lead to new legal challenges or a change in how AI training is regulated remains to be seen.