Key Points

• The UK Competition and Markets Authority (CMA) has asked Meta to explain changes to its commercial terms amid concerns that they could permit the company to scrape publishers’ websites.
• CMA chief executive Sarah Cardell told lawmakers on 20 October 2026 that the regulator would meet Meta on 21 October as it considers potential competition enforcement.
• Meta removed the word “train” from its revised terms and narrowed the stated purposes for using crawled material to advertising and business products.
• Legal experts and publishers remain concerned that the remaining wording, including “develop”, could still permit broad use of website content.
• Meta says the changes clarify existing terms and do not change how it uses web crawlers. It also maintains that website owners can communicate their preferences through robots.txt.
• The revised commercial terms are scheduled to take effect on 30 October 2026, putting pressure on publishers to assess the implications for their business accounts.
• The News Media Association has criticised the terms, while Movement for an Open Web is preparing a challenge involving the CMA and European Commission.
• The dispute comes as UK regulators are already addressing how major technology platforms use publishers’ material in artificial intelligence services.

The UK Competition and Markets Authority (CMA) is examining concerns surrounding Meta’s revised website-crawling terms after publishers and legal experts warned that the provisions could allow the technology company to collect and reuse material from websites associated with business accounts. The regulator has asked Meta to explain the changes and is considering possible competition enforcement, with CMA chief executive Sarah Cardell telling lawmakers on 20 October 2026 that the watchdog would meet the company the following day. Meta says its revisions clarify the scope of its commercial terms rather than change its web-crawling practices, but questions remain about how the wording could affect publishers’ control over their content and its potential use in artificial intelligence products.

The dispute has developed rapidly ahead of the planned implementation of Meta’s revised commercial terms on 30 October 2026. Although the company has removed an explicit reference to training from the controversial wording, critics argue that the remaining language may still be broad enough to cover some AI-related uses of collected material.

In its report published on 20 October, MLex reported that the CMA had asked Meta to explain the changes as it considered potential enforcement action. The report also said the regulator had identified Meta as a potential candidate for designation under the UK’s digital markets regime. This indicates regulatory scrutiny, rather than a final finding that Meta has breached competition law.

The controversy places the interests of publishers, the commercial policies of a major technology platform and the CMA’s digital markets responsibilities at the centre of a debate about access to online content.

Why has the CMA raised concerns about Meta’s website-scraping terms?

The issue arose after Meta introduced revised commercial terms that appeared to give the company broad permission to access websites associated with businesses using its products and services.

In a report published on 17 October 2026, Anthony Ha of TechCrunch explained that the original wording allowed Meta to use content from specified business domains, webpages and URLs to provide, develop, improve, research, train and maintain the integrity of its products and services. Legal experts interviewed by the publication warned that the language could potentially be interpreted as permission to collect content from publishers’ websites and use it to train AI models.

The concern was not simply that Meta operates web crawlers. Automated website collection is used across the technology industry for purposes including search, data retrieval and service development. The controversy instead centred on whether the terms gave Meta broader contractual permission to collect content associated with business accounts and what it could subsequently do with that material.

According to Andrew Deck and Hanaa’ Tameez of Nieman Journalism Lab, the original wording raised concerns that Meta could scrape websites associated with business accounts and potentially use the collected material for AI training. Their report said the company subsequently removed the word “train” and narrowed the wording to refer to advertising and business products.

The CMA’s request for an explanation brings the issue into the UK’s competition-policy debate. However, the regulator’s scrutiny should not be confused with a determination that the wording is unlawful or that Meta has used it to collect content in breach of publishers’ rights.

What changes has Meta made to its commercial terms?

Meta revised the disputed wording following criticism from publishers and legal experts. The most visible change was the removal of the explicit reference to training, alongside language narrowing the stated purposes for using collected content to advertising and business products.

As reported by TechCrunch on 17 October, the original provision referred to using content to “provide, develop, improve, research, train, and maintain the integrity of” Meta’s products and services. The revised wording removed the explicit training reference and clarified the connection with advertising and business products.

Meta’s position is that the revision clarifies the intended scope of terms that had been consolidated for its commercial products. The company says the change does not reflect a new approach to web crawling.

Critics argue that removing the word “train” does not necessarily resolve the issue. The remaining term “develop”, for example, could be interpreted broadly, potentially leaving uncertainty about whether some AI-related processing would fall within the permitted purposes.

In a report published on 17 October, Dealroom News described this concern and noted that legal experts questioned whether the amended terms adequately limited the possible use of crawled material. The report also highlighted criticism that the wording did not clearly settle the question of how website owners could prevent unwanted collection.

The central disagreement is therefore about the practical and legal scope of the revised terms, not merely the removal of a single word. Meta says its clarification reflects existing practice; critics say the remaining language still needs to be assessed carefully.

What has Meta said in response to publishers’ concerns?

Meta has rejected the suggestion that the revised terms represent a change in its web-crawling practices.

According to Anthony Ha of TechCrunch, a Meta spokesperson said the consolidated crawling terms applied only to the company’s advertising and business products and that the wording had been clarified to make that scope more explicit. The spokesperson also said the changes did not alter how Meta used web crawlers.

Meta has also maintained that website owners can communicate their preferences through robots.txt, a standard file used to give automated crawlers instructions about which parts of a website they should access.

In its coverage of the controversy, MLex reported on 19 October that a Meta spokesperson said the terms were not intended to authorise the scraping of news websites or establish new rights overriding website owners’ preferences.

The company’s explanation is an important part of the dispute. A change in contractual language does not, by itself, prove that a company has changed its technical practices or that it has accessed material unlawfully.

Nevertheless, publishers and legal commentators have questioned whether the revised wording provides sufficient clarity about the scope of collection, permitted uses and the practical protections available to website operators. Those questions remain relevant even where a company says its underlying practices have not changed.

How are UK publishers and media organisations responding?

Publishers have raised concerns that broad crawling provisions could affect their control over original reporting and other website content. The issue is particularly significant for news organisations, which invest in journalism but may also depend on technology platforms to distribute articles and reach audiences.

As reported by Resultsense, relaying Press Gazette coverage on 20 October, Theo Bamber, chief executive of the News Media Association, criticised Meta’s approach and called on ministers and regulators to protect British creators. The report said the association had expressed concern about the precedent that the terms could establish for publishers.

The same report said Movement for an Open Web, a campaign group advocating changes to the digital publishing ecosystem, was preparing a legal challenge through the CMA and the European Commission. Its director, James Rosewell, argued that the potential consequences of incorporating publishers’ content into AI models could be difficult to reverse.

The campaign group has urged publishers to consider deleting their Meta business accounts before the revised terms take effect. This is a campaign recommendation, not a regulatory requirement. The available reporting does not establish that all publishers must leave Meta’s platforms or that deleting an account is the only way to protect website content.

These responses illustrate the difficult position facing publishers. Meta’s platforms can be important channels for reaching readers, but businesses must also assess the contractual conditions attached to their use and the potential implications for their content.

What role does robots.txt play in the dispute?

The robots.txt file allows website operators to publish instructions for automated crawlers. Website owners can use it to communicate which sections of their sites particular automated agents should or should not access.

However, the mechanism is not an absolute technical barrier. Its effectiveness depends on whether crawlers recognise and respect the instructions, and it does not resolve every question about copyright, contractual permissions or other legal rights.

Meta says website owners can continue to use this mechanism to communicate their preferences. Critics, however, have questioned whether the revised terms make those protections sufficiently clear and whether the contractual language adequately addresses the potential uses of collected material.

For publishers, the distinction between expressing a preference and having an enforceable remedy is important. A technical instruction may help communicate an access policy, but whether a particular collection or reuse of material is lawful depends on the facts and applicable law.

The debate also highlights the importance of transparency about crawler identification, the purpose of data collection and the relationship between technical access and commercial reuse.

What could the CMA do next?

The CMA’s immediate reported step is to seek an explanation from Meta and discuss the issue with the company. According to MLex, Sarah Cardell told lawmakers on 20 October that the regulator would meet Meta on 21 October while considering possible competition enforcement. The report also identified Meta as a potential candidate for designation under the UK’s digital markets regime.

These developments do not mean that the CMA has already concluded that Meta breached the law. The outcome will depend on the regulator’s assessment of the facts, the applicable legal framework and the company’s responses.

The digital markets regime gives the CMA powers to address competition concerns involving major digital firms that meet the relevant criteria. Potential designation is a separate step from establishing a breach arising from a particular contractual provision.

The regulator may need to consider the extent of Meta’s market position, the practical effects of the terms on publishers and businesses, and whether the company’s policies create barriers or disadvantages for other market participants. No final determination on these matters is established by the reporting currently available.

The situation also demonstrates why businesses should distinguish between a regulator requesting information, considering designation and taking formal enforcement action. Each stage has different implications for the company and the wider market.

How does the dispute relate to the UK’s wider regulation of AI?

The Meta controversy comes as UK regulators are addressing the use of publishers’ content in AI-powered services offered by major technology companies.

On 3 June 2026, the CMA announced measures requiring Google to provide publishers with effective ways to prevent their content from being used in AI-powered search features, including AI Overviews. The measures also addressed attribution of publishers’ content in AI-generated results and publishers’ ability to opt out of having their material used to fine-tune AI models.

That action concerns Google and should not be treated as a ruling against Meta. It nevertheless demonstrates the regulator’s broader interest in the relationship between digital platforms, publishers’ content and competition.

The underlying issues include whether publishers can exercise meaningful control over their material, whether users can identify the sources of information presented by AI services and whether businesses have fair opportunities to negotiate the commercial use of their work.

For companies developing or deploying AI systems, these questions reinforce the importance of understanding data provenance, contractual permissions and the potential risks of relying on material collected from third-party websites. The Artificial Intelligence (AI) and Legal, Contracts and Procurement fields are directly relevant to understanding these governance and commercial considerations.

What should publishers and businesses consider before 30 October?

With the revised terms scheduled to take effect on 30 October 2026, publishers and businesses using Meta’s commercial products have a limited period in which to review the changes and assess their implications.

Businesses may wish to examine the wording applicable to their accounts, identify which websites and domains are associated with their products or services, review their crawler preferences and seek specialist advice where the legal consequences are uncertain.

Publishers should also distinguish between the terms that govern their use of Meta’s products and any separate legal rights they may have over their content. The available reports do not establish that every business account will be affected in the same way or that leaving Meta is necessary in every case.

For technology companies, the controversy is a reminder that broad contractual language can create reputational and regulatory risks even when the company says it is intended to clarify existing practices. For publishers, it underscores the importance of understanding how digital distribution arrangements interact with the protection and commercial value of original content.

The CMA’s engagement with Meta may provide further clarity about the regulator’s concerns, but the outcome of the reported meeting and any subsequent action should not be assumed in advance.

What happens next in the Meta scraping controversy?

The immediate milestones are the reported CMA–Meta meeting on 21 October and the planned commencement of the revised commercial terms on 30 October 2026. The regulator’s assessment, Meta’s explanations and any further response from publishers or campaign groups will help determine how the dispute develops.

At this stage, the positions remain distinct. Meta says the wording changes clarify the scope of its commercial products and do not change its web-crawling practices. Publishers and legal experts have questioned whether the remaining terms provide sufficient safeguards and whether they could permit broad use of content associated with business accounts.

The CMA is reported to be examining the issue and considering potential competition enforcement, but no final finding against Meta has been established in the cited reporting.

The wider significance extends beyond a single clause in a commercial agreement. As AI systems increasingly rely on large quantities of information, technology companies, publishers and regulators face continuing questions about transparency, access, commercial rights and accountability.