Germany : The High Court of Düsseldorf sets out criteria to determine whether an AI image may be copyrighted
Link to German decision (in German)
Logos generated by AI do not have copyright protection
In a landmark German ruling, the Munich Local Court decided that AI-generated logos require human creative input to be copyrighted (case no. 142 C 9786/25).
The court clarified that copyright protection under German law depends on whether the AI-generated result bears the creative imprint of the human author. Human contribution may arise during or after the process.
The court emphasised that human influence must be clear and shape the outcome. In this case, prompts were considered generic, so creative decisions were attributed to AI algorithms.
The court ruled that the logos did not adequately express the plaintiff's personality and denied them copyright.
The decision is important for companies, designers and legal professionals as it establishes when AI-generated works can be copyrighted.
See the decision below (26.02.2026)
https://www.gesetze-bayern.de/Content/Document/Y-300-Z-BECKRS-B-2026-N-1513?hl=true
LIKE COMPANY v. GOOGLE
Opinion published: the opinion of the European Copyright Society on case C-210/21 (Like Company v Google).
The hashtag European Copyright Society has published its comment on the request for a preliminary ruling in case C-250/25 (Like Company v Google).
This reference from the Budapest Regional Court in the Like Company v Google case (Case C-250/25) is considered the first opportunity for the Court of Justice of the European Union (CJEU) to rule on an issue at the heart of artificial intelligence disputes around the world. Do AI models and their results constitute an infringement of exclusive intellectual property rights?
ecs-comment-on-the-like-company-reference.pdf
What the case is about:
First AI Case Before the CJEU: Like Company v. Google Ireland (C-250/25)
Danish media industry takes OpenAI to court
The case centres on OpenAI’s use of content from Danish press publications to develop and provide its artificial intelligence services, most notably ChatGPT. This includes the reproduction of works, which occurs when OpenAI collects and uses data to train its large language models (LLMs), as well as the subsequent reproduction and distribution of content from Danish press publications through ChatGPT outputs.
A united Danish media industry takes OpenAI to court – DPCMO
RTI and MEDUSA FILM vs PERPLEXITY AI
Reti Televisive Italiane (RTI) and Medusa Film have filed a lawsuit in the Civil Court of Rome against Perplexity AI Inc., a US company active in generative artificial intelligence. This is the first legal action in Italy aimed at challenging copyright infringements related to the use of artificial intelligence. According to the two Mediaset Group companies, Perplexity has used numerous audiovisual and cinematographic contents owned by them without permission and on a large scale to train its artificial intelligence systems. This activity, according to RTI and Medusa, constitutes a violation of copyright and other related rights. With this appeal, RTI and Medusa are asking the Court to recognise the illegality of the conduct, block any further unauthorised use of their content and order Perplexity AI to pay damages. They are also requesting the application of a daily penalty in the event of any future violations.
Link to the PR: Comunicato Stampa_2467.pdf
Comment:
GETTY IMAGES
The UK High Court has granted Getty Images permission to appeal against parts of the copyright ruling in Stability AI's favour. This follows the November ruling that largely favoured Stability AI on copyright grounds, leaving only limited trademark issues outstanding.
KNESCHKE V. LAION
On 10th December 2025 the Hanseatic High Court dismissed the claim of case of Robert Kneschke v. LAION.
Appeal to the German Supreme Court was allowed because, in view of the increasing use of artificial intelligence, the legal issues raised in this case have significance beyond the specific case.
Not only has the Court dismissed the case but it has also reviewed the interpretation of what "machine-readable" means.
We will publish comments here.
🔗 AI Training: Plain language reservation of use not sufficient for OLG Hamburg | heise online
🔗 Higher Regional Court Hamburg Confirms AI Training was Permitted (Kneschke v. LAION) - Bird & Bird
GEMA v. OPEN AI
GEMA, German collecting society for music, has won against Open AI.""Both the memorisation in the language models and the reproduction of the song lyrics in the chatbot's outputs constitute infringements of copyright law," the court ruled in a case brought by the German music body GEMA."
News
🔗 German music body wins copyright case against OpenAI
Press statement of the Bavarian Justice
🔗 Pressemitteilung 11/2025 - Bayerisches Staatsministerium der Justiz
"Both the memorisation in the language models and the reproduction of the song lyrics in the chatbot's outputs would constitute an infringement of copyright exploitation rights. These are not covered by any restrictions, in particular the restriction for text and data mining."
🔗 GEMA v OpenAI: Memory is Fragile. Garbage Lasts Forever. – Open Future
GETTY IMAGES v. STABILITY AI
Getty Images claimed that Stability AI’s image-generation model, Stable Diffusion, had scraped millions of its copyrighted photographs, illustrations, and videos without a licence. They also claimed that Stability AI had used these images to train the model and that the model’s outputs reproduced substantial parts of Getty Images’ works. Getty also claimed infringement of its database rights (with regard to its image collection), trademark infringement (as some outputs bore Getty watermarks/trademarks) and passing off (as this misled users into thinking the images were from or licensed by Getty).
The court largely ruled in favour of Stability AI on the remaining copyright claims. The judge found that, since the Stable Diffusion model did not store or reproduce Getty Images’ works in a way that constituted an infringing 'copy', and since the training and outputs did not amount to the reproduction of a substantial part of the protected works in the UK, there was no copyright infringement under UK law. law.
Regarding trademark infringement, Getty achieved a partial victory: the court found that some generated images bore Getty’s watermarks/trademarks in a way that could constitute trademark infringement. However, the judgment described this success as 'historic and extremely limited in scope'.
🔗 High Court Judgment Template
Getty Images's Statement of 4th November may be found here:
🔗 Getty Images issues statement on ruling in Stability AI UK litigation - Getty Images
Background articles:
🔗 Getty Images v Stability AI: A landmark judgment reinforcing the need for the UK government to amend its copyright laws | Kluwer Copyright Blog
Key statement: "The decision of Mrs Justice Smith DBE will apply to diffusion models and models that behave like Stable Diffusion but not necessarily other AI models (...) The issue of secondary infringement is not yet fully settled as this case only dealt with one type of AI model. AI companies should therefore pay attention as, if their models actually keep works in their memory, under UK copyright law they could be infringing under s. 22-23 of the CDPA."
Previous articles:
Getty Images
🔗 Getty Images Drops Main Copyright Claims Against Stability AI in UK Legal Case | PetaPixel
MACHINE-READABLE OPT-OUT
On 22 October, a Danish court ruled that the standard plain-text rights reservation language included in a privacy policy satisfied the requirements for rights reservation under Article 4 (see link below). The policy was accessible via a link in the homepage footer and was available in HTML format.
🔗 Domsoversigt BS-42485/2025-SHR
LIKE COMPANY v. GOOGLE
The case was originally brought by Like Company Kft (the Hungarian publisher) as the applicant, against Google Ireland as the defendant. Muhami+3EUR-Lex+3ddg.fr+3
The referring court is the Budapest Környéki Törvényszék (Budapest regional court), which lodged the request for a preliminary ruling to the Court of Justice of the European Union (CJEU) on 3 April 2025. Digital Policy Alert+2ddg.fr+2
The key factual allegations: Like Company claims Google’s chatbot (Gemini) reproduced and made available portions of its protected press-publisher content without authorisation
The Hungarian court asked the CJEU to interpret EU law (Request for preliminary ruling under Article 267 of the Treaty on the Functioning of the European Union (TFEU)
The referring court asked questions about:
The key underlying question is whether the activity of a large online platform (Google Ireland) using press-publisher content in an AI system amounts to a copyright-relevant act under Article 15 of the DSM Directive?
Articles on the case:
From the law firm:
🔗 CJEU to rule on landmark AI and copyright case from Hungary
🔗 Copyright, AI, and the Future of Internet Search before the CJEU
🔗 First AI Case Before the CJEU: Like Company v. Google Ireland (C-250/25)
Request for preliminary ruling here:
🔗 AI CEPIC AI WG Submissions and Documents – Google Drive
The development of the case at the CJEU may be followed here :
🔗 CURIA - Case information