The News/Media Alliance today sent Meta chief executive Mark Zuckerberg a three-page letter demanding that the company withdraw its revised Commercial Terms before they take effect on October 30, 2026, and asking Meta to respond publicly by October 26. The Arlington, Virginia trade association, which says it represents more than 2,000 newspaper, magazine and digital media publishers in the United States, argues that the crawling provision in the terms amounts to an exercise of market power, and states that its members "do not consent" to the terms and "reserve all of their legal rights."

In Short

Meta changed the rulebook for every business that uses Facebook, Instagram, WhatsApp or Messenger, and the new rules let Meta send robots to those businesses' websites to collect what is there. The biggest US news publishers' trade group says that means newspapers and magazines would have to hand over their journalism for free just to keep promoting it on Meta's apps, and it has asked Meta to scrap the clause and reply by October 26. Meta says the rules are about ads and business products, not news sites, so the dispute now turns on whether Meta changes the wording before October 30 or the publishers take formal action.

What the letter demands

The letter, dated October 22, 2026, is signed by Danielle Coffey, president and chief executive of the News/Media Alliance, and is headed "An Open Letter to Mark Zuckerberg: Meta's New Terms Take Publishers' Journalism Without Permission or Payment." It opens with a single demand: that Meta "withdraw its revised Commercial Terms before their claimed effective date of October 30, 2026."

The alliance's own press release, published the same day under a staff byline, narrows that request. It "calls on Meta to withdraw the crawling provision from its updated terms of use," rather than the whole document. The letter itself never separates the crawling language from the rest of the rewrite, though nearly all of its argument concerns Section 1.

The core objection is about scope. According to the letter, the new terms "appear to require every business that uses Facebook, Instagram, WhatsApp, or Messenger through a business account to grant Meta extraordinary and disproportionate privileges to exploit the content of its websites for free." Publishers, the alliance says, "would have no way to decline this, short of abandoning their use of Meta's business products and forfeiting access to Meta's massive audience."

Then comes the competitive argument, which is the letter's sharpest claim. "News publishers will be forced into providing their valuable journalism to support the development of products that may be used to compete with them for readers and advertisers, under duress and without compensation," it states. A few lines later: "Meta's unilateral conduct, given the scale and reach of its social media platforms, is an exercise of market power and is anticompetitive. Meta's behavior risks undermining the robust licensing market for publishers' content and could allow Meta to simply snatch that content for free."

Those are the alliance's characterisations. No court or regulator has found that the terms are anticompetitive, and the letter does not say that any complaint has been filed with an antitrust authority.

How the alliance reads Section 1

PPC Land reported on October 16 that Section 1 of the revised terms authorises Meta to send automated software to domains associated with a business's products or services and to create test credentials to reach them. The letter quotes that passage in the same form: Meta may use "web scrapers, bots, robots, spiders, crawlers, and user agents and create test credentials to access, index, cache, analyze, collect, retrieve or crawl" any domain "associated with your products or services, or that are made available through your or users' use of Meta Company Products," and return periodically for updates.

The alliance draws its widest conclusion from the second half of that clause. "The claimed license appears to reach every article on a publisher's website or that a reader shares, in addition to content a publisher posts to a Meta platform or pages a publisher promotes there," the letter says. The reference to articles "that a reader shares" rests on the words "users' use of Meta Company Products," a prong that, on its face, is not limited to the business's own domains.

On test credentials, the letter goes a step beyond the text. It states that test credentials "could let Meta register its own accounts to get past registration walls, and Meta has not said whether that includes paywalls." The terms do not mention registration walls or paywalls; the letter's footnote cites Nieman Lab's report that Meta declined to say whether test credentials would be used to bypass paywalls.

The "train" deletion and what replaced it

The letter confirms a sequence that PPC Land has been tracking for a week. "After publishers objected, Meta deleted the word 'train' from the initially released version, amongst the litany of uses to which publishers must consent, and now says it will use the content to 'develop, improve, research, and maintain advertising and business products,'" according to the alliance. Its source for the revised wording is a Nieman Lab article by Andrew Deck and Hanaa' Tameez dated October 17, 2026.

That revised wording matches the clause as Adweek quoted it today in the report on Meta making advertisers solely responsible for its AI ad edits. It does not match the UK English edition of Meta's preview page that PPC Land checked on the evening of October 18, which still contained the "train" sentence word for word. Neither the letter nor the press release says when the edit reached Meta's published text, or whether every language edition carries it.

For the alliance, the timing matters less than the substance. "The revision changes little, and the terms remain deeply vague," the letter states. "'Advertising and business products' is undefined, and the phrase 'develop…advertising and business products' is, as one law professor put it, 'still broad enough to drive a truck through.'" A footnote identifies the professor as James Grimmelmann, quoted by Nieman Lab. The press release frames the same point as legal experts saying the revised terms "still grant Meta overly-broad scraping privileges without explicit limitations."

The press release describes Meta's weekend changes as "minor revisions to language." October 17, 2026, the date of the Nieman Lab article the letter cites, fell on a Saturday.

A second licence, and the indemnity

The letter also points to a separate sentence in Section 1 that governs content a business shares with Meta. Under that language, Meta may use the content to "provide, develop, improve, research, and maintain Meta Company Products." According to the alliance, "Meta Company Products" is broadly defined "to include the full suite of Meta websites, products, and apps, current or future, including AI products such as Meta AI or the Muse AI agent, and the Business Agents tools that the revised terms have incorporated by reference." The letter cites Meta's own definitions page, facebook.com/legal/meta-company-products, for that reading.

Footnote 5 adds a point the main text does not develop. "In Section 6a, publishers are required to indemnify Meta for any third-party claims brought by third parties with regard to the use of such content on the Meta Company Products," it reads. The alliance's conclusion: "Not only does Meta force publishers to make an impossible warranty but also require them to hold Meta harmless for any claims." The warranty in question is the representation, in Section 1, that a business holds the rights needed for Meta's uses. For a publisher, those rights can depend on freelance contracts, wire service agreements and photo licences that were never written with a platform's AI products in mind.

Acceptance, the letter notes, is passive. Meta "claims that continued use counts as acceptance, and a business that objects must 'stop all access or use of our Products for business or commercial purposes,'" according to the alliance, citing Section 7.

Why blocking is not the answer, according to the letter

Meta's public defence has rested on robots.txt. The alliance addresses it directly and offers three objections.

The first is textual. "Meta says publishers can still block its crawlers with robots.txt, but Meta's terms do not confirm that it will abide by the restrictions, nor do they promise that Meta will limit scraping to identified bots," the letter says. That matches PPC Land's reading of the earlier text: the terms neither carve robots.txt out nor override it.

The second is practical, and it is the argument most specific to publishers that advertise on Meta. "Blocking is also not a practical option: publishers generally must allow Meta's ads crawler for their advertising to function, and the terms do not limit how content collected by that crawler may be used," according to the letter. A publisher that promotes articles or subscription offers through Facebook and Instagram ads needs Meta's systems to fetch the landing pages. Shutting the door on Meta's bots would affect that traffic.

That concern is familiar from the wider crawler debate. Meta already runs a mixed-use crawler, Meta-ExternalAgent, that uses one robots.txt token for both AI training and indexing, so a site cannot permit one purpose while refusing the other. The letter applies the same logic to advertising: if the crawler a publisher must admit is also covered by a broad licence, the technical control does not separate the uses.

The third objection concerns existing deals. The letter states that Meta does not "confirm that these terms do not supersede existing agreements." Publishers with negotiated licensing or content arrangements with Meta would want to know which document prevails. The revised terms, as quoted in the letter, do not say.

The press release condenses the second and third objections into one line, quoting the letter: "Publishers cannot practically disallow or block Meta's crawlers without harming their ability to promote content as a business account user, which would limit user access to quality journalism content on the platform."

The money at stake

The letter puts a figure on the dispute. "If the terms take effect on October 30, they purport to grant Meta the right to exploit, without prior permission or fair compensation, journalism for which, by one estimate, Meta already owes U.S. publishers nearly $2 billion a year," it says.

The estimate is not the alliance's own. The footnote cites "Paying for News: What Google and Meta Owe US Publishers," a 2024 Columbia University working paper by Haaris Mateen and co-authors, which estimated that Meta owes US publishers $1.9 billion a year, "before accounting for AI." The figure is an academic model of what fair compensation for news on Meta's platforms might look like, not a sum any court has awarded or Meta has acknowledged.

The press release turns the number into a broader claim. "Meta harvesting billions of dollars worth of journalism for free would cause severe harm to news publishers by irreversibly crippling the multibillion-dollar market for licensing news to AI developers," the release says. It gives no figure for the size of that licensing market.

The letter's harm section is written in the language of a pre-litigation notice. "Publishers that cannot capture the value of their reporting will produce less of it: fewer reporters covering fewer communities, fewer investigations, and less original news for the public," it says. "This harm will fall hardest on the local and independent publishers least able to leave Meta's platforms." The final sentence of the section reads: "Once Meta has copied publishers' content and built it into its products, the harm will be practically impossible to undo."

Why stress irreversibility? In US litigation, irreparable harm is one of the conditions for emergency relief such as a preliminary injunction. The letter does not threaten a lawsuit, but its wording lays the groundwork that such a request would need.

Four conditions for future talks

The letter does not reject crawling outright. It asks Meta to withdraw the terms and "hold discussions around future changes with NMA and publishers," and it lists what those discussions should cover:

  • the specific Meta Company Products and uses for which Meta wants to use publishers' scraped content;
  • how publishers can opt in to permit scraping, "separate from obtaining access to Meta's business products";
  • the specific crawlers, fetchers or other bots that will be used;
  • how Meta will ensure that publishers' robots.txt files, registration walls and paywalls are respected.

The second point is the structural one. An opt-in that sits apart from business account access would undo the bundling the alliance objects to. The fourth would turn Meta's verbal assurances about robots.txt into contractual commitments.

The letter then uses Meta's own statements as leverage. "Meta has publicly stated that the crawling terms 'only applied to our ads and business products' and are 'not about scraping news sites.' If so, it should have no objection to putting future commitments into the terms themselves, as an outcome of these discussions," it says. It also rejects Meta's description of the update. "Meta has said that the terms update changes nothing. However, that statement is not supported by the language at issue, which is a clear change over the prior terms." The previous Commercial Terms, effective March 6, 2024, contained no crawling clause, as PPC Land's October 16 comparison found.

The final section is the part with potential legal weight. "Because the terms may take effect on October 30, we ask Meta to respond publicly by October 26, 2026," the letter says. "In the meantime, this letter serves as notice that, in the event Meta adopts these revised terms over our members' express objections, our members do not consent to them and reserve all of their legal rights."

It goes further. Meta, the alliance writes, "does not have permission or the right to use our members' content for any purpose unless they have expressly, in writing, authorized such use." Members' presence on Meta's platforms "does not and cannot grant Meta the right to scrape, access, or use their owned content from their websites." The letter closes: "Put simply, any access or use of our members' content for AI purposes without their express written permission is contrary to applicable law."

Can a trade association withhold consent on behalf of its members to a contract each member accepts individually by continuing to use a product? The letter does not address that question, and the Commercial Terms bind each business separately. The notice is best read as a record of objection that individual publishers may later rely on, rather than as a substitute for their own decisions about whether to keep using Meta's business tools after October 30.

Who the alliance is

The News/Media Alliance describes itself as a nonprofit headquartered in the Washington, D.C. area and formerly the Newspaper Association of America. Its history spans about a century of mergers between publishing associations, most recently with MPA-The Association of Magazine Media in 2022. The letter states that its membership "accounts for nearly 90 percent of the daily newspaper circulation in the United States and over 500 individual magazine and digital-only brands."

Its own materials do not agree on the headline figure. The letter and the press release say "over 2,000" publishers. The alliance's About Us page says its members represent "over 2,200 diverse publishers in the United States."

The group has been one of the most active voices against unlicensed use of journalism by technology companies. In May 2025, Coffey called Google's AI Mode "the definition of theft". In April 2026, the alliance wrote to Common Crawl demanding that it remove publisher content on request and bar AI training uses. And in September, the alliance and America's Newspapers brought more than 300 publishing executives to Washington to lobby for the Stealth Bot Prohibition Act, which would require AI scraping bots to identify themselves and disclose their purpose. That bill, which the alliance wrote, would fine undisclosed AI crawlers $53,000 per violation.

Its use of antitrust language against a platform is not new either. In December 2024 the alliance asked the Department of Justice and the Federal Trade Commission to investigate Google's site reputation abuse policy, citing Section 2 of the Sherman Act and Section 5 of the FTC Act.

Why it matters for publishers and marketers

The letter shifts the Meta dispute from interpretation to positions on the record. Until today, the argument over Section 1 ran through journalists, legal academics and Meta spokespeople. Now the largest US news trade body has formally told Meta that its members object, set a date for a public answer and laid out what it would accept instead.

The bundling argument has parallels elsewhere. Penske Media has argued in federal court that Google's "forced choice" between appearing in search and supplying content for AI features breaks a longstanding bargain with publishers. The alliance's letter applies a similar theory to a different gate: access to Meta's advertising and business tools rather than search rankings.

Outside the United States, regulators have already pushed Meta on payment for news. France's competition authority in July ordered Meta to resume negotiations with press publishers within 15 days, and in May the EU Court of Justice upheld Italy's rules requiring platforms to offer press publishers fair compensation. The US has no equivalent statute, which is one reason the letter leans on competition language and on a notice of non-consent.

For advertisers outside publishing, the stakes are less direct but not absent. The crawling clause applies to every business using Meta's commercial products, not only news organisations. A retailer whose product pages Meta crawls for ad recommendations is covered by the same text the alliance objects to. If Meta writes narrower commitments into the terms in response to publishers, those limits would apply to all business users.

Meta has four days to answer by the alliance's deadline, and eight before the terms apply. Neither the press release nor the letter records any response from Meta to the letter itself.

Timeline

Summary

Who: The News/Media Alliance, a US trade association for more than 2,000 newspaper, magazine and digital publishers, led by president and chief executive Danielle Coffey; Meta Platforms and its chief executive Mark Zuckerberg; and every publisher and business that uses Meta's commercial products.

What: An open letter demanding that Meta withdraw its revised Commercial Terms, or at least their crawling provision, calling the clause anticompetitive, arguing that blocking Meta's crawlers is impractical for publishers that advertise on its platforms, setting four conditions for future talks and serving notice that members do not consent to the terms.

When: The letter and press release are dated today, October 22, 2026. The alliance asks for a public response by October 26, 2026. The terms take effect on October 30, 2026.

Where: The letter was issued from Arlington, Virginia, and addressed to Meta in Menlo Park, California. It concerns use of Facebook, Instagram, WhatsApp and Messenger business accounts by US publishers.

Why: The alliance says the terms would let Meta collect publishers' journalism without permission or payment and use it to build products that compete with them, while Meta says the crawling language applies only to ads and business products. Meta's removal of the word "train" did not settle the dispute, because, in the alliance's view, "advertising and business products" remains undefined.