Meta has removed the word “train” from its updated commercial terms of service, a change that will take effect on October 30, after a wave of criticism from legal experts and media organisations.
The move follows last week’s backlash over the company’s new commercial policy, which some analysts said could allow Meta to scrape content from websites linked on business pages without paying licensing fees. By requiring business users to share large amounts of data, the policy could effectively provide Meta with free AI‑training material.
Legal experts warned that the policy’s wording might grant Meta “permission to endlessly scrape websites linked on business pages,” a loophole that could let the company avoid millions of dollars in licensing costs from news outlets.
A Meta spokesperson confirmed that the company had not notified users about the language change after the backlash, a fact that could add to confusion ahead of the policy’s rollout.
In a statement to Ars, Meta acknowledged Nieman Lab’s initial report but did not address the follow‑up article that reiterated the experts’ concerns about the policy’s implications for AI training and content licensing.
<small>Source: Ars Technica — read the original story there.</small>
