Meta removes language that sparked scraping and AI training concerns
meta training
| Source: TechCrunch | Original article
Meta has removed language from its terms that had raised concerns about web scraping and AI training, though the company says the change does not alter how it crawls the web.
Meta has quietly revised its user‑terms documents, stripping out the clause that sparked debate over whether the company could scrape web content for artificial‑intelligence training. The change was confirmed by a Meta spokesperson, who said the updated language “does not reflect a change in the way Meta crawls the web.”
The amendment follows criticism that the previous wording could be interpreted as granting Meta carte blanche to harvest articles, images and other publicly available material for its AI models without compensation to publishers. By removing the contentious phrasing, Meta appears to be responding to pressure from media organisations and regulators who have warned that unrestricted data harvesting could breach emerging European and U.S. AI‑related legislation.
As we reported on 16 October, Meta’s new terms had raised the prospect that the firm could train AI on news sites for free, a move that many feared would tilt the competitive landscape in favour of the tech giant. The latest edit does not alter the company’s crawling practices, according to the spokesperson, but it does signal a willingness to adjust contractual language in the face of public scrutiny.
Stakeholders will be watching whether the revision satisfies calls for greater transparency and whether it forestalls potential legal challenges. Regulators in the EU and the United States are expected to continue probing the adequacy of platform terms for AI data use. Further updates from Meta, especially any clarification of its data‑use policies or new engagement with news publishers, will be key indicators of how the debate over AI training data evolves.
Sources
Back to AIPULSEN