Meta Backs Down as Publishers Revolt Over AI Scraping Terms
Meta Backs Down as Publishers Revolt Over AI Scraping Terms
Meta’s proposed rewrite of its business terms was meant to take effect on October 30. Instead, it set off an unusually broad backlash from publishers, legal experts and regulators who feared that keeping a Facebook business page could mean giving the company access to crawl any external website linked to it for AI development.
The central objection was not merely that Meta might collect more data, but that publishers would have little practical choice. News organisations with Facebook pages could not opt out while retaining those pages, critics said; the proposed rules also required test credentials that could allow Meta’s bots to get past paywalls. Lawyers warned the vague wording could potentially let the company disregard robots.txt instructions, the standard mechanism website owners use to limit automated crawling.
Pressure mounted quickly. Courts in Brazil and Norway issued preliminary injunctions after legal challenges, while the UK Competition and Markets Authority opened an investigation into whether Meta was using a dominant position to acquire valuable AI-training material. In the US, Senators Adam Schiff and Marsha Blackburn also urged Mark Zuckerberg to withdraw the terms.
Publishers viewed Meta’s attempted reassurances with deep suspicion. Theo Bamber, chief executive of the UK News Media Association, called the company’s posture “staggering” and asked: “Does Meta think we can’t read?” He argued that swapping the word “train” for “develop” did not meaningfully narrow what Meta could do with scraped content.
The October 30 deadline was first moved to November 20. Then, on October 28, Meta communications vice-president Andy Stone said the company had heard publishers’ objections and would not make changes without “a full consultation with industry groups and leaders.” Meta says existing terms and crawling practices remain unchanged, and that content owners can still use robots.txt. But critics note that its revised language still does not explicitly commit the company to respecting those preferences.
The retreat followed an outcry that the rules could enable Meta to scrape large volumes of news-site data and potentially use it to build competing products. For now, the proposed overhaul is delayed indefinitely; the fight over who controls the open web’s data is not.
Write a comment