🔍 Read the full analysis: The EU Court Adviser’s AI Copyright View, And Why It Matters on ThorstenMeyerAI.com
Get smart everyday buys delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
On Nov. 3, 2026, Advocate General Maciej Szpunar advised the EU Court of Justice that chatbots retrieving and reusing news articles may infringe press publishers’ rights in certain circumstances. He said the example involving Google’s Gemini does not appear to meet that test; the opinion is nonbinding, and the court has not set a judgment date.
Advocate General Maciej Szpunar advised the EU Court of Justice on Nov. 3 that AI chatbots which retrieve news articles and reuse them in answers may, in some circumstances, infringe press publishers’ rights. In the dispute over Google’s Gemini and a Hungarian publisher, however, Szpunar said the conditions for infringement do not appear to be met, leaving the final decision to the court.
The opinion concerns Case C‑250/25, Like Company v Google Ireland, which the court describes as the first generative AI case to reach it. Like Company, which operates Hungarian news sites, brought proceedings in Hungary over Gemini’s use of its articles. The company alleges that Google used the articles to train its language models and reproduced and made them available through chatbot answers between June 13, 2023, and May 31, 2024, according to the court’s press release.
Szpunar’s proposed test focuses on how a chatbot obtains and presents an article. An answer may infringe if the service accessed the publication unlawfully, such as by bypassing a paywall or an anti-scraping measure, or if it reproduces the article’s distinctive expression so closely that a user can forgo the original publication. Repeating the same facts, without reproducing protected expression, would not by itself be enough, the opinion says.
Applied to the example before the court, Szpunar considered those conditions apparently absent: the article was not behind an access barrier, the sample answer did not appear to reproduce recognisable wording, and Google appeared to have safeguards. Those are conclusions in an advisory opinion, not a ruling on liability. The Hungarian court will assess the case in light of the EU court’s eventual interpretation.
Infographic 01 · Where the line runs
When a chatbot answer infringes a publisher's right
Advocate General Szpunar's test in Case C‑250/25 for chatbots that retrieve news articles in real time and reuse them in their answers.
Stays free
- Reporting the same facts. Nobody owns the news itself (pts 199–200).
- A short factual summary, presented differently (pt 208).
- An overview built from several sources, restructured, with links back (pts 208, 234).
- Individual words or very short extracts (pt 201).
Can infringe
- Getting around a paywall or anti-scraping block to fetch the article (pt 188).
- An answer that keeps the article’s structure and phrasing, even paraphrased, so readers can skip the original (pts 202, 205, 207).
- Systematically recognisable articles in answers to related questions: making available to the public (pts 213, 240).
The test, step by step
For a chatbot that fetches a news article live (RAG) and answers a user
- Is it an online use by a service in the EU?Yes for RAG answers to EU users, wherever the servers are (pts 165–169).
- Did the bot get in lawfully?No circumvention of effective technical measures (pt 188).
- Is the article’s form of expression recognisable in the answer, to the point of substitution?Side-by-side comparison; facts do not count (pts 196–209).
- Is the provider responsible?If inherent to its business model, or no reasonable measures such as respecting blocks and output filters (pts 221–222).
- Temporary copies during retrieval are then exempt under Art. 5(1) InfoSoc; the TDM exception does not apply to RAG (pts 179–189).
An Advocate General’s opinion, not a judgment. General information, not legal advice.
Sources: Opinion of AG Szpunar, C‑250/25 (provisional text), CURIA press release No 155/26 (3 Nov 2026). Point numbers refer to the provisional text.
Powered by Thorsten Meyer AI
The Line Between News and Copying
The opinion could shape how AI search and chatbot services use publishers’ material in Europe. If the Court of Justice adopts Szpunar’s approach, services may need to account for access controls and avoid answers that closely reproduce the distinctive wording of an article. That could affect product design, sourcing practices and negotiations over licenses.
For publishers, the proposed standard offers a possible route to challenge certain forms of reuse, but it does not give them control over facts reported in their stories. The distinction matters because a chatbot can summarize what happened without necessarily copying how a publication expressed it. The opinion therefore points toward a boundary between reporting shared information and substituting for a particular article through recognizable reproduction.
Users could also be affected if services respond to legal risk by relying more on multiple sources, providing links, or changing how they quote and summarize pages. Those are possible consequences, not changes announced by Google or directed by the adviser. The court’s final judgment will determine whether Szpunar’s proposed interpretation becomes the relevant EU standard.
Products Worth Considering
Samsung 27-inch Odyssey G5 QHD Gaming Monitor (G51F)
Dell XPS 13 9350 Copilot+ Laptop, 13.4-inch 120Hz Display, Intel Core Ultra 7 256V, 16GB RAM, 512GB SSD, Ubuntu Linux 24.04 LTS
Apple Mac mini (2026) with M5 Pro chip and AppleCare+ (3‑year)
As an affiliate, we earn on qualifying purchases.
How the Gemini Dispute Reached Luxembourg
Like Company filed its case in a Budapest court under the EU press publishers’ right, established by Article 15 of the 2019 Copyright in the Digital Single Market Directive. The right was designed in part to let publishers control or license certain online uses of extracts, including by news aggregation services. The dispute asks whether that framework also applies when a chatbot retrieves an article to respond to a user.
The example at the center involved a Hungarian article published in July 2023 about a singer’s reported proposal to bring dolphins to Lake Balaton. A user asked Gemini to summarize the article, and the chatbot returned a summary in Hungarian. The Hungarian court referred four questions to Luxembourg: two concerning AI training and two concerning chatbot answers. The EU hearing took place on March 10, 2026, after written observations from nine member states and the European Commission.
On model training, Szpunar’s main recommendation was procedural: he proposed that the court declare the two training questions inadmissible because the case record did not establish that Like Company’s articles were used to train Gemini. He reasoned that the article’s publication date and Gemini’s release timing made that alleged use implausible, and that the example appeared to involve live retrieval from the web. He added, as a separate point, that training is generally covered in principle by the EU’s text-and-data-mining exception, subject to rights holders’ ability to opt out in a machine-readable form. That discussion was not a finding that these articles were used for training.
“the first generative AI case to reach the Court of Justice”
— Court of Justice of the European Union
What the Opinion Does Not Decide
Szpunar’s opinion is not binding. The Court of Justice may follow it, modify its reasoning or take a different approach, and no date for the judgment has been set. The opinion also does not decide whether Google infringed Like Company’s rights; the national proceedings remain relevant to that question.
The record, as described in the opinion, did not establish that Like Company’s articles were included in Gemini’s training data. The training discussion should not be read as a factual finding about what data Google used. It is also unclear how the proposed test would apply across different chatbot designs, types of access restriction, or answers that combine wording from several publications.
Awaiting the Court’s Final Judgment
The Court of Justice will issue its judgment after considering the Advocate General’s opinion. The opinion says preliminary references took an average of 16.9 months in 2025, according to court statistics, but that historical average does not establish when this case will be decided. Once the EU court rules on the referred questions, the Hungarian court will continue the dispute using the interpretation it provides.
Until then, the key development is guidance proposed by an independent adviser, not a final legal requirement. Publishers, AI services and users will need to wait for the court’s judgment to learn whether the proposed distinction between shared facts and recognisable expression is adopted.
Key Questions
Has the EU Court of Justice ruled against Google?
No. Szpunar issued a nonbinding opinion on Nov. 3, 2026. The Court of Justice has yet to give its judgment, and the opinion does not decide Google’s liability.
When could a chatbot answer infringe publishers’ rights?
Under the adviser’s proposed test, infringement may arise if a chatbot accesses an article unlawfully, such as by bypassing a paywall or anti-scraping measure, or reproduces its distinctive expression so closely that users can forgo the original. Repeating the underlying facts alone would not be enough under that approach.
Did the opinion find that Gemini was trained on Like Company’s articles?
No. Szpunar said the record did not support the claim that the articles were used for training and recommended treating the training questions as inadmissible. The example instead appeared to involve live web retrieval for a chatbot answer.
What happens next in the case?
The Court of Justice will deliver a final judgment, with no date announced in the source material. The Hungarian court will then handle the national dispute in light of the EU court’s interpretation.
Source: ThorstenMeyerAI.com
