Seven trade organizations representing news and magazine publishers in Europe and Canada announced a transatlantic coalition on October 26, 2026, and asked the European Union and Canada to include in the joint declaration of the 21st EU-Canada Summit, which takes place in Montreal on October 29 and 30, a "firm commitment to framework conditions supporting free, independent, and sustainable media, as well as trusted professional journalism."

The members are the European Newspaper Publishers' Association (ENPA), the Alliance de la presse d'information générale (APIG), the European Magazine Media Association (EMMA), the European Publishers Council (EPC), News Media Europe, News Media Canada and Magazines Canada. Their statement singles out artificial intelligence, and in particular AI-generated summaries embedded in search engines, as the most immediate challenge for the sector. It does not name a single technology company. Nor does it cite a traffic figure, a revenue figure or any draft legal text. It was distributed through Business Wire in English and French under a Brussels and Montreal dateline, at 6:00 PM Eastern Daylight Time, and it is an advocacy position from organizations with a direct commercial stake in the outcome. No response from the European Commission or the Government of Canada appears in it.

In Short

Groups that speak for thousands of newspapers and magazines in Europe and Canada have teamed up and asked their governments to promise, at a meeting in Montreal on October 29 and 30, to protect independent news. They say AI tools, especially the summaries that now sit at the top of search results, use their articles without asking or paying and keep readers from clicking through to their sites. That, they say, costs them readers, subscriptions and advertising money. What they want are rules and tools that let publishers decide who uses their work, see how it is used and get paid for it.

Seven groups, two continents

According to the release, the seven organizations "represent many other national and regional associations," and the coalition speaks for "thousands of trusted news and magazine titles across both sides of the Atlantic." It gives no count of titles, no list of national affiliates and no combined audience or circulation figure.

The membership mixes levels. Four of the groups - ENPA, EMMA, the EPC and News Media Europe - work at European level. APIG is a French national association for the general information press. News Media Canada and Magazines Canada are national Canadian bodies.

Two of the members arrive with active cases against the largest platforms. The EPC, together with Corint Media, filed complaints that preceded the European Commission's formal antitrust investigation into Google's use of publisher content for AI, opened on December 9, 2025 under Article 102 of the Treaty on the Functioning of the European Union. On October 23, 2026, the same council asked the Commission to halt Meta's revised crawling terms before November 20, arguing that Meta's removal of the word "train" from its crawling clause did not exclude AI training, fine-tuning or AI-generated answers.

APIG has been just as busy in Paris. Its complaint, alongside one from the Société des Droits Voisins de la Presse, led France's Autorité de la concurrence to order Meta on July 8, 2026 to resume good-faith negotiations and hand over remuneration data within 15 days, in interim decisions 26-MC-01 and 26-MC-02. About a month later, on August 11, 2026, the alliance filed a referral with the same authority against Google, 20 days after Google switched on AI Overviews and AI Mode in France on July 22, 2026.

The release itself mentions none of these cases. Its single direct quotation is attributed jointly to the coalition members:

"Our news publications and magazines share a common mission: to inform, enlighten, and contribute to public debate. That work has a cost, and it has value. When platforms appropriate our content without authorization or compensation, it is not only publishers that are weakened, but our democracies as well. The Montreal Summit is an opportunity for Europe and Canada to demonstrate that they are committed to defending, together, independent and sustainable media."

What the coalition is asking for

The core request is narrow in form and broad in substance. Leaders are asked to put a commitment to media "framework conditions" into a summit declaration. The release does not define the phrase, and it does not say whether the coalition has proposed wording to either delegation.

Beyond the declaration, the coalition sets out three positions. First, it "insists that legislation must now evolve at the pace of technological change, particularly artificial intelligence (AI), and be reinforced through stronger safeguards that provide publishers with meaningful tools to authorize, monitor, and enforce the use of their content." The release does not identify which statutes should change, in which jurisdiction, or how.

Second, it says the coalition "will work collectively to foster a transparent AI ecosystem that protects intellectual property, facilitates licensing, and ensures that publishers receive fair compensation while retaining full control over their content and the value it generates." That combination - licensing on one side, full control on the other - resembles a condition the EU's top court attached to national publisher-pay rules in May, discussed below.

Third, and most technical, it asks that "transparency requirements should also extend to downstream data usage rights." That is a request about what happens to content after it has been collected, not only about whether a crawler may collect it in the first place. The release ties this point to a specific industry standard, the content telemetry specification developed by the SPUR Coalition.

The case against AI summaries

The release makes three linked claims. It says that "in the absence of appropriate rules and transparency in the enforcement of rights, unidentified AI bots and agents are exploiting press content at scale without authorization or compensation." It says that "these AI-generated summaries divert readers away from press publications' websites and appropriate value created by publishers, resulting in substantial losses in digital audiences and the subscription and advertising revenues that support them." And it argues that "these losses undermine a key pillar of publishers' business models and weaken their capacity to produce reliable, diverse, and public-interest journalism."

None of those claims comes with a number. Figures do exist elsewhere, though they come from different methods and markets. French media regulator Arcom has estimated traffic losses of 33% to 38% from AI-generated summaries on European markets where those services operate, and a randomized study of 1,065 desktop Chrome users found that AI Overviews cut outbound organic clicks by 39.8%. The Reuters Institute's 2026 trends survey of 280 executives in 51 countries found publishers expect search traffic to fall a further 43% over three years, while Chartbeat data put the year-over-year decline in Google Search traffic to publishers globally at 33% between November 2024 and November 2025.

Google disputes the premise. When the Independent Publishers Alliance, Movement for an Open Web and Foxglove filed an EU antitrust complaint over AI Overviews on June 30, 2025, Google responded that many of the traffic claims rested on highly incomplete and skewed data. The coalition's release does not engage with that argument.

The reference to "unidentified AI bots and agents" has a close American parallel. More than 300 publishing executives gathered by the News/Media Alliance and America's Newspapers asked Congress in September to pass the Stealth Bot Prohibition Act, introduced in July 2026, which would require AI scraping bots to identify themselves and state their purpose. TollBit counted more than 22 billion AI bot scrapes in the first half of 2026. The dominant opt-out mechanism, robots.txt, only works when a crawler announces itself and honors the file, which explains why identification sits so high on the publishers' list.

The point is not theoretical for this coalition. When Meta delayed its revised Commercial Terms and Self-Serve Ad Terms from October 30 to November 20, 2026, a Meta spokesperson told Reuters that content owners could still use robots.txt to signal crawling preferences. The EPC's position, set out in its October 23 letter, is that robots.txt is no substitute for enforceable contractual limits.

The rights the coalition wants enforced

The release frames its demands as protecting existing law rather than inventing new rights. "In response to mounting challenges, both the European Union and Canada have introduced frameworks to protect the free press, including the neighbouring right for press publishers, as well as Canada's Online News Act," it says. Each framework has a different track record.

The neighbouring right in Europe

The neighbouring right for press publishers comes from Article 15 of EU Directive 2019/790 of April 17, 2019, which France transposed through its Act of July 24, 2019. Its scope was clarified on May 12, 2026, when a 13-judge Grand Chamber of the Court of Justice of the European Union ruled in case C-797/23, a dispute between Meta Platforms Ireland and Italy's communications regulator AGCOM. The court held that Article 15 does not prevent national rules granting publishers a right to fair remuneration, including rules that require platforms to negotiate, supply data and accept a regulator-set amount when no deal is reached. It attached conditions. Publishers must keep the ability to refuse authorization or to grant it free of charge, and platforms that do not use press publications cannot face a payment duty.

France has been the most active enforcer. The Autorité fined Google 250 million euros in March 2024 in a neighbouring rights dispute. Meta's agreement with the Société des Droits Voisins de la Presse expired on December 31, 2024 and its agreement with APIG on January 31, 2025, and members had received no remuneration since.

Canada's Online News Act

Canada's Online News Act received Royal Assent on June 22, 2023, and its implementing regulations came into force on December 19, 2023. On October 28, 2024, the Canadian Radio-television and Telecommunications Commission approved Google's exemption application, under which Google pays 100 million Canadian dollars a year to the Canadian Journalism Collective in exchange for a five-year exemption from individual bargaining.

The release presents the Act as a framework that protects the press. It does not mention that Meta took a different route. Meta began ending news availability on Facebook and Instagram for all users in Canada on August 1, 2023, in response to the Act, according to Global News, and argued at the time that links to news made up less than 3% of content in users' feeds. Canada's law has therefore produced one large annual payment from one platform and a news withdrawal by another, and the coalition's statement addresses neither outcome directly.

The AI Act

Under Article 53 of the EU AI Act, providers of general-purpose AI models have since August 2, 2025 been required to prepare technical documentation, a copyright policy and a public summary of their training content, with August 2, 2027 set as the deadline for models placed on the market before August 2, 2025. The release offers that law a cautious endorsement: "The EU's AI Act also constitutes a recognition of the need for lawfulness, accountability and transparency in the age of AI." The Commission has sent requests for information to companies that have not published their training summaries, according to Reuters.

Those obligations concern how models are built. The coalition's demand for transparency on "downstream data usage rights" reaches further, into retrieval, citation and display of content inside AI products after training has finished. That is the territory the SPUR standard is designed to cover.

SPUR and content telemetry

The release describes the SPUR Coalition's content telemetry standard as one of "the initiatives advanced by the industry" and says it "offers a practical foundation for interoperable accountability mechanisms." It adds that the standard "provides publishers with the information needed to continue innovating, adapt their business models, and make informed licensing decisions."

SPUR - Standards for Publisher Usage Rights - was formed in March 2026. Its members include The Guardian, the Financial Times, the BBC, Sky, The Times of London, MediaHaus and the Associated Press, which joined as its first US founding member alongside what were then thirty publisher members and six affiliate members. Its technical working group announced a draft telemetry standard on June 12, with public comment open until July 24. The draft breaks AI-content interaction into five tracked events: retrieved, grounded, cited, displayed and engaged. The "grounded" event refers to grounding, in which an AI system anchors a generated answer to a specific retrieved source.

Version 1.0 of the standard was released on October 2, 2026. It covers multimodal content, not only text, and is meant to sit alongside existing C2PA provenance work. SPUR invited OpenAI, Anthropic, Google, Meta and Microsoft to join an AI licensing advisory board of around twenty members, with a first technical committee meeting scheduled for October 15. Pilot programs with technology and AI companies are the next step, though no participants or dates had been named by early October. SPUR's technical lead, Alex Springer, said OpenAI had not commented on the standard, though the company was aware of it.

Two gaps stand out. The release does not say whether any of the seven coalition members belongs to SPUR. And it does not say whether any AI company has agreed to emit the telemetry the standard describes. A reporting standard depends on the systems that generate the events, and those systems belong to the companies the coalition says are appropriating its content.

Other industry efforts have taken different routes. IAB Europe's "Crawling for Compensation" framework, published in September 2025, focused on access controls, discovery protocols and cost-per-crawl monetization APIs rather than post-ingest reporting. Bilateral licensing has also grown: Google said on October 15, 2026 that more than 400 publications now take part in its News AI licensing program, without disclosing payment amounts or terms. Both sit outside the coalition's statement, which mentions licensing only in general terms.

The Montreal summit

The summit date in the release matches the Council of the European Union's own record. A provisional agenda for the Permanent Representatives Committee, dated October 22, 2026, lists an exchange of views on the EU-Canada summit in Montreal on October 29-30, 2026. The release does not say whether either delegation has seen or accepted language on media, and it offers no fallback if the declaration omits it.

The timing places the coalition's call between two dates that matter to publishers. Meta's revised terms are now due to take effect on November 20, 2026. In the UK, which sits outside both jurisdictions named in the release, the main obligations of the Competition and Markets Authority's first conduct requirement on Google take legal effect on December 3, 2026, with page-level grounding controls due on March 3, 2027. Google has already introduced a domain-level Search Console toggle, rolling out first to a subset of UK site owners, that lets them keep content out of AI Overviews, AI Mode and AI Overviews in Discover; Google says opting out does not affect conventional rankings. That is the closest existing example of the "meaningful tools to authorize, monitor, and enforce" the coalition describes, and it comes from a regulator in neither the EU nor Canada.

Litigation in the United States has offered publishers less. On October 1, 2026, Judge Amit Mehta dismissed Penske Media's antitrust suit over AI Overviews, finding that Penske had not pleaded an actual agreement under which Google promised it traffic.

Why this matters for the marketing community

The release ties AI summaries directly to "subscription and advertising revenues." For media buyers, that is not an abstract policy debate. Publisher audiences are the inventory that open-web campaigns buy, and the same AI search surfaces that publishers say divert readers are themselves advertising placements: AI Overviews began showing ads in May 2024.

The dispute also reaches beyond news sites. Meta's crawling clause, the subject of the EPC's latest letter, applies to business websites generally, which means the terms publishers are contesting would also govern how Meta crawls advertisers' own sites. Any EU action prompted by publisher complaints would therefore land on a contract that marketers accept as well.

If AI companies adopt SPUR's standard, publishers would be told when their content is retrieved, used for grounding, cited, displayed and engaged with - usage data that, in the release's words, would inform "licensing decisions." Whether governments in Montreal treat that as a matter for a summit declaration, and whether any AI company agrees to report it, remains open.

Timeline

Summary

Who: Seven publisher trade organizations - the European Newspaper Publishers' Association, the Alliance de la presse d'information générale, the European Magazine Media Association, the European Publishers Council, News Media Europe, News Media Canada and Magazines Canada - representing thousands of news and magazine titles, addressing the European Union and the Government of Canada.

What: The groups formed a transatlantic coalition and asked for a "firm commitment" to framework conditions for free, independent and sustainable media in the summit's joint declaration, stronger safeguards that let publishers authorize, monitor and enforce use of their content by AI systems, transparency over downstream data usage rights, and they endorsed the SPUR Coalition's content telemetry standard.

When: The coalition was announced on October 26, 2026, ahead of the 21st EU-Canada Summit on October 29 and 30, 2026.

Where: The release carried a Brussels and Montreal dateline; the summit takes place in Montreal.

Why: The coalition says AI-generated summaries in search engines and unidentified AI bots use press content without authorization or compensation, diverting readers and cutting the subscription and advertising revenue that funds journalism. The release provides no figures for those losses and names no company.