Meta Platforms has revised its commercial terms of service, effective October 30, to eliminate a clause that appeared to grant the company broad permission to scrape content from business-associated websites for AI training. The removed sentence stated, "Meta may use any content from such domains, web pages, or URLs to provide, develop, improve, research, train, and maintain the integrity of, the Meta Company Products, services, and features provided under these Commercial Terms and any other applicable supplemental terms." This change comes after legal scholars and media outlets scrutinized the updated terms, which were initially perceived to allow Meta to access valuable publisher content without separate licensing agreements. The initial language, published earlier this month, had raised alarms among legal experts and news organizations. Cornell law professor James Grimmelmann, for example, warned that the provision could compel newsrooms to permit Meta to use their original reporting for AI training without compensation, or risk losing their audience on Meta's platforms. Other legal professionals also noted that the clause could give Meta access to material that would likely cost millions of dollars in a licensing market. The terms apply to any business that uses Meta's products for commercial purposes, including those managing a Facebook Page, running ads, or utilizing WhatsApp for business. In response to the growing criticism, Meta communications executives, including Andy Stone and Daniel Roberts, clarified the company's position. Stone stated on X that the summary of the report was false and that the update was intended to ensure Meta could pull accurate product information for on-platform ad and content recommendations, not to scrape news sites. Roberts added that Meta was not changing how its web crawlers operate and that traditional robots.txt files would continue to protect paywalled articles and private content from being scraped. Meta's existing ad terms already allow it to crawl advertisers' websites for uses including training, but the October update extended these provisions to businesses not directly purchasing ads. Despite Meta's assurances and the removal of the explicit "train" language, some skepticism remains within the digital media community. Representatives from the News/Media Alliance indicated that while the removal of the specific word was a positive step, the updated terms still grant Meta broad crawling permissions without explicit contractual guarantees regarding content usage. Legal experts have also questioned whether the updated agreement could eventually be interpreted by Meta to claim contractual permission to collect material even when website owners have attempted to block crawlers. Meta has previously stated that its AI, including Llama-based features, is trained on a combination of publicly available data, licensed data, and information from its own products and services, such as publicly shared Facebook and Instagram posts. The company has also faced copyright litigation concerning the use of pirated book datasets for AI training. The updated commercial terms, even with the revised language, still authorize Meta to use automated software, including web scrapers and bots, to access and collect content from domains and pages associated with a user's products or services. Businesses that do not agree to the updated terms are instructed to stop using Meta's products for business or commercial purposes.