Meta to delay new commercial terms criticised by publishers, advertisers

Meta is delaying the effective date of two updated rule sets: its Commercial Terms and its Self-Serve Ad Terms. These terms had been scheduled to take effect on October 30. The change matters because publishers and digital advertisers believe the terms could give Meta broad rights to use their content for AI services.

The new date is 20 November. Before then, Meta says it will explain what is changing and what is not. It will also describe how its web crawlers work and clarify how content owners can use robots.txt to signal preferences about access.

The delay may help Meta address complaints before regulators decide whether to act. The European Commission welcomed the move. However, the European Publishers Council still wants regulators to suspend the terms while they investigate. Meta has rejected the claim that the terms provide unlimited content rights.

Meta’s Commercial Terms and Self-Serve Ad Terms are the updated contractual rules governing commercial arrangements with Meta, including self-serve advertising. The article does not provide their full text or define every provision. It shows that the rules matter because they may determine what Meta can do with content connected to business pages, advertising, and other services.

Publishers and digital advertisers complained that the terms would give Meta an unlimited right to use their content for AI services. Meta rejected that claim. Meta also said it would explain what is changing, what is not, and how its web crawlers operate.

The dispute is therefore about contractual reach, not just advertising administration. The European Publishers Council says clear safeguards are missing against unrelated exploitation of publishers’ content. It argues that specific services should not automatically create broad rights over a publisher’s website or archive.

The complaint concerns the possible breadth of Meta’s claimed rights. Publishers and digital advertisers say the updated terms would give Meta an unlimited right to use their content for its AI services. That would reach beyond the particular business page or advertising service a publisher chooses to use, according to the criticism.

The European Publishers Council gave a direct example. It said maintaining a business page or advertising a subscription should not confer general rights over a publisher’s website and archive. In its view, using one Meta service should not automatically cover unrelated material elsewhere on the publisher’s site.

Meta rejects the claims, so the alleged scope remains disputed. The council wants regulators to suspend the terms while they investigate. Its preferred safeguard is separate, express authorisation from the relevant rightsholder for uses beyond those necessary to provide the specific service requested.

robots.txt is a plain text file placed by a website owner to communicate instructions or preferences to web crawlers and AI bots. Crawlers are automated systems that access web pages. In this dispute, the file is important because it offers publishers a way to signal how they want automated systems to approach their content.

For example, a content owner can use robots.txt to communicate preferences about crawler or bot access to parts of a website. The article does not list the file’s specific commands or guarantee how every system will respond. It does say Meta will explain its use of web crawlers and make clear that content owners can still use robots.txt.

That clarification is part of Meta’s response to complaints about the updated terms. It may help partners understand what the terms change and what remains available to them. The European Publishers Council nevertheless says contractual safeguards are still unclear.

The European Commission acts as the EU’s antitrust watchdog in this dispute. It welcomed Meta’s decision to delay the terms. The Commission also said it had sent requests for information to companies that had not published summaries about the data used to train their AI models.

UK antitrust authorities, together with EU authorities, have received complaints about Meta’s updated commercial terms. Publishers and digital advertisers say the terms would give Meta an unlimited right to use their content for AI services. Meta has rejected that interpretation.

The European Publishers Council and other groups are pressing regulators to suspend the terms while they investigate. The council says contractual safeguards against unrelated exploitation are unclear. Its Executive Director, Angela Mills Wade, argues that uses beyond a requested service should require separate, express authorisation from the relevant rightsholder.

Meta says the delay is meant to create more time for explanations, not that it has already changed the substance of the terms. The updated Commercial Terms and Self-Serve Ad Terms will now take effect on 20 November rather than October 30. Meta will discuss what is changing, what is not, and how its crawlers operate.

That extra time could help Meta respond to publishers and advertisers’ concerns. It could also give the European Commission and UK antitrust authorities more time to consider complaints and information about AI data practices. The article says the decision could stave off potential regulatory action, but it does not predict a final regulatory outcome.

The terms could still face opposition. The European Publishers Council wants regulators to suspend them while investigating. Meta has not announced a rewrite in the article. The immediate next step is explanation and consultation before the new effective date.

The conflict exists because publishers treat their articles as copyrighted content, while AI companies use such material to train large language models. Publishers worldwide are increasing their challenge to that practice when it happens without permission. Their concern is that valuable editorial work may be used for AI services without clear authorisation or safeguards.

AI companies, as the article describes, cite fair-use arguments. That position treats some unlicensed uses as potentially legally permitted rather than automatically forbidden. The article does not resolve whether fair use applies to Meta’s terms or to any particular training activity. It presents the issue as a clash between copyright owners and companies defending their practices.

Meta’s updated terms have intensified that dispute because publishers and advertisers say they could give Meta an unlimited right to use content for AI services. The European Publishers Council wants separate, express authorisation for uses beyond a requested service. Regulators are examining related concerns.