🔍 Read the full analysis: Why Facts And Writing Matter In The AI Chatbot Copyright Debate on ThorstenMeyerAI.com
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
TL;DR
On Nov. 3, 2026, Advocate General Maciej Szpunar advised the EU Court of Justice that chatbot answers using online news articles may infringe press publishers’ rights in certain circumstances. His proposed test focuses on unlawful access or recognizable reproduction of an article’s expression, not simply the reuse of its facts. The nonbinding opinion says the conditions do not appear to be met in the Gemini case, which remains for a Hungarian court to decide.
Advocate General Maciej Szpunar advised the European Union Court of Justice on Nov. 3 that chatbots which retrieve online news articles and reuse them in answers may infringe press publishers’ rights if they obtain material unlawfully or reproduce its recognizable expression. The opinion in Case C‑250/25, Like Company v Google Ireland, is not a binding judgment, and Szpunar said the proposed test does not appear to be met in the dispute over Google’s Gemini chatbot and a Hungarian article about dolphins.
The case concerns a Hungarian article published in July 2023 about a singer’s reported plan to bring dolphins to Lake Balaton. Like Company, which operates Hungarian news portals, sued Google, alleging that its articles were used to train Gemini and were reproduced through chatbot answers between June 13, 2023, and May 31, 2024. The dispute reached the EU court after a Budapest court referred four legal questions to Luxembourg.
Szpunar’s proposed distinction is between facts and expression. A chatbot does not infringe the publishers’ right merely by reporting the facts in an article, he said. The risk arises if the service accessed an article by circumventing a paywall or anti-scraping measure, or if its answer preserves the article’s distinctive wording or form so closely that a reader can do without the original. The opinion says that, in the example before the court, there appear to have been no access barriers and no recognizable wording in the answer.
The opinion treats the training allegations separately. Szpunar advised that the questions about training should be dismissed as inadmissible because the record did not show that Like Company’s articles were used to train Gemini. He also said that training on dedicated computers would not count as the kind of online use covered by the EU press publishers’ right. His further discussion of text-and-data-mining rules was offered for completeness, not as a ruling on the facts of the case.
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
A Boundary Between Facts and Copying
The opinion offers publishers and AI companies a possible legal framework for disputes over chatbot answers: the facts in reporting remain free to convey, while a sufficiently close reuse of an article’s expression may trigger the press publishers’ right. That distinction could shape how AI search and chatbot services handle news they retrieve from the web, including whether answers should summarize across sources, avoid close paraphrase, or link readers to the original publication.
For publishers, the proposed test could support claims when a service bypasses access controls or substitutes for reading an article by reproducing its distinctive expression. It does not give publishers ownership of the underlying news or a general veto over summaries of reported facts. For users, the eventual judgment could affect how directly chatbots answer questions about current reporting and how prominently they identify or link to sources.
These are potential implications, not settled rules. The Advocate General’s opinion is advisory; the Court of Justice will issue the binding judgment, and the Hungarian court will then resolve the underlying dispute under the applicable law.
Popular right now
Apple AirPods Max 2 Wireless Over-Ear Headphones – Midnight
KitchenAid Artisan 5-Quart Tilt-Head Stand Mixer, Almond Cream
Dyson Car+Boat Cordless Handheld Vacuum
As an affiliate, we earn on qualifying purchases.
How the Gemini Dispute Reached Luxembourg
EU law has granted press publishers a related right since 2019 under Article 15 of the Copyright in the Digital Single Market Directive. The right was developed with online uses such as news aggregation in view. The Like Company dispute asks whether it also reaches generative chatbot answers that draw on live web results.
According to the Court of Justice’s press release, Like Company alleged both that Google used its articles in Gemini’s training data and that the chatbot repeatedly reproduced them. The Hungarian court referred two questions about training and two about chatbot answers. Nine EU governments and the European Commission submitted written observations, and the hearing took place on March 10, 2026.
On training, Szpunar said the record did not support the claim that the dolphin article was in Gemini’s training data. It was published in July 2023, while Gemini 1.0 was released in December 2023; the opinion reasons that training and data collection generally occur in advance. It says the answer could instead have come from retrieval-augmented generation, in which a chatbot retrieves web pages at the time of a query and generates a response from them. The user had included the site’s domain in the prompt, according to the opinion.
The Binding Ruling Is Still Pending
The Court of Justice has not issued its judgment, and the Advocate General’s opinion does not decide the case. The court may follow the proposed reasoning, modify it, or reach a different conclusion. No judgment date has been announced.
The record described in the opinion also leaves the training allegation unresolved on its merits: Szpunar considered the questions inadmissible because evidence did not establish that the articles were used for training. His additional discussion of the EU text-and-data-mining exception is not a finding that Google trained Gemini on Like Company’s material or that any particular use was lawful.
The eventual legal and practical reach of the test is also unknown. The opinion identifies access barriers and recognizable expression as relevant circumstances, but the court’s judgment will determine how the publishers’ right applies. The Hungarian court must still assess the dispute using the facts and the final EU interpretation.
Court Judgment and Hungarian Proceedings
The next major step is a binding judgment from the Court of Justice. The court has not given a date. Its 2025 statistics put the average time for preliminary references at 16.9 months, but that figure does not establish when this case will be decided.
After the EU court answers the Hungarian court’s questions, the Budapest court will continue the case and decide the parties’ dispute. Until then, Szpunar’s opinion provides guidance rather than a final determination of Google’s liability or the legality of any specific chatbot answer.
Key Questions
Did the EU Court of Justice rule against Google?
No. Advocate General Szpunar issued a nonbinding opinion. The Court of Justice has yet to deliver its judgment, and the Hungarian court will decide the underlying dispute afterward.
When could a chatbot answer infringe publishers’ rights?
Under the opinion’s proposed test, possible infringement arises if a chatbot obtains an article by bypassing access barriers or reproduces its recognizable expression closely enough that users can dispense with the original. Reporting the same facts alone is not enough, according to Szpunar.
Did Szpunar find that Gemini was trained on the articles?
No. He advised that the training questions were inadmissible because the case record did not show that Like Company’s articles were used in Gemini’s training data. His separate discussion of text-and-data-mining rules was not a finding about Google’s actual training data.
What happens next in the case?
The Court of Justice must issue its binding response to the Hungarian court’s questions. The Budapest court will then continue the case and determine the dispute; no date has been set for the EU court’s judgment.
Source: ThorstenMeyerAI.com
