Meta delays new business terms after publishers object to AI scraping language

Meta has postponed new business terms that publishers say would let it scrape their websites and use the content to train AI. The terms were due to take effect on October 30. They will now apply from November 20, a Meta spokesperson told The Verge and Reuters.

The delay followed complaints from US publishers, a letter to Mark Zuckerberg and appeals to regulators in the European Union and the United Kingdom. It does not settle the dispute. Meta says it needs more time to explain the terms. It has not said it will withdraw them.

What the terms say

The update covers Meta's Commercial Terms and Self-Serve Ad Terms. According to Reuters, the spokesperson said: "We are extending the effective date of our updated Commercial Terms and Self-Serve Ad Terms to 20 November."

The Verge reports that the terms appeared to allow Meta to scrape the sites of businesses that use its platforms freely, and to use that material for AI training. Meta later revised the text to remove language about "training" its products. Legal experts argued that this change does not rule out the possibility.

Meta disputes how critics read the terms. Reuters reports that publishers and digital advertisers told EU and UK antitrust authorities the terms would give Meta an unlimited right to use their content for its AI services. Meta rejected those claims.

The publishers' objections

The News/Media Alliance is a trade group of more than 2,000 newspaper, magazine and digital publishers. On October 22 it sent a letter to Zuckerberg demanding that Meta withdraw the revised Commercial Terms, and in particular the crawling provision, before they took effect. The Verge reports the letter warned of "severe harm" to publishers.

The Alliance makes three arguments. First, the terms would let Meta scrape content from any website associated with a business account on its platforms, and use it for business products that include AI tools. Second, publishers "cannot practically disallow or block Meta's crawlers without harming their ability to promote content" on Meta's platforms. Third, the Alliance says this raises antitrust concerns, because Meta would be using its position in social media to extract concessions in the market for news licensing. The letter also cites an estimate that Meta already owes US publishers nearly $2 billion a year for their journalism.

In Europe, digital advertisers and publishers in the EU and the UK asked regulators to make Meta halt the new terms, The Verge reports. Reuters says the complaints went to antitrust authorities in both jurisdictions. Neither report names the specific agencies.

Meta's position: robots.txt

Meta spokesperson Daniel Roberts told The Verge the delay is meant "to allow more time for us to explain to our partners and stakeholders what's changing and what's not." He said Meta will use the time to explain "how Meta uses web crawlers and the fact that content owners will still be able to use robots.txt to signal preferences."

Robots.txt is a file on a website that tells automated crawlers which pages they may visit. Following it is voluntary for the crawler. Meta's argument is that publishers keep control because they can use the file to keep its crawlers out.

The Alliance's letter targets that argument directly. A publisher that relies on Meta to promote its articles cannot easily block the crawler that makes the promotion work. If one crawler does both jobs, the choice in robots.txt becomes all or nothing: accept the crawling, or lose some of the distribution.

What it means for work

The dispute is over who controls journalism once it has been published, and on what terms it can be reused to build AI products. For reporters and editors, the question is whether work paid for by a newsroom becomes input for another company's models without a separate licence. Publishers' business and legal staff now have to read platform terms of service as licensing documents and decide case by case whether distribution on Meta is worth the access it grants. The delay only buys time: Meta has not said it will change the crawling provision. It is also unclear whether EU or UK authorities will act, or whether robots.txt counts as meaningful consent when the same crawler serves both promotion and data collection. The outcome will affect what news licensing deals are worth, and so how much revenue is left to pay for the work itself.