GERMANY

COUNTRY CHAPTER

2026 RANKINGS

Artificial intelligence (AI) has become standard in everyday working life. The use of AI is growing rapidly, particularly for the creation of content such as photos, texts, films and music. From a legal perspective, this raises the question of whether or not rights can be established in relation to the use of AI. In particular, this relates to whether the described content is eligible for copyright protection.

The question of copyrightability is relevant in many respects. When a person creates content mentioned above, copyright and/or ancillary copyrights are regularly established for works, which in turn can be licensed for a fee. In addition, third parties may be prohibited from using the works. If the content is not eligible for protection, then the possibility of licensing and the right to prohibit its use are ruled out.

A prompt entered into AI by a person can also result in photos, etc. In this case, the question arises as to whether the person can also create protectable works through this prompt if the output is otherwise generated by the AI. The US Copyright Office and certain courts rejected the copyrightability in general (see US Copyright Office report, dated 29 January 2025, „Copyright and Artificial Intelligence”, Part 2: Copyrightability, at 18; District Court for the District of Columbia, judgment of 18 August 2023, case-no. 22-1564 - Thaler v. Perlmutter (denial of certiorari by Supreme Court, judgment of 2 March 2026, case-no. 25-449). Recent case law takes a more nuanced approach to this issue and sees possibilities for prompts to create protectable works.

The Munich Local Court recently had to rule on the copyrightability of AI-generated logos. In principle, the court assumes that a work exists if, despite the software-controlled process flow of the AI, human creative influence is still exerted on the design of the output (see Munich Local Court, judgment of 13 February 2026, case-no. 142 C 9786/25). The decisive factor in prompting is that the creative abilities of the prompter are expressed independently in the output "by making free and creative decisions and thus also giving the output his personal touch“. If the creative decision is left to the AI by means of only general, open-ended instructions, this is not sufficient, even if these prompts are numerous and thus gradually change the appearance of the output.

In the specific case, the court rejected the copyrightability of the logos because the prompts contain only general instructions that did not justify any creative influence. The court did not consider that the following prompts would have a decisive influence on the design of the AI output: "Design an original, abstract logo", "The design should be modern, minimal, and distinctly original, with clear evidence of creative interpretation", "Style: Clean flat design with custom geometric abstraction".

In conclusion, it can be said that the copyrightability of AI-generated content cannot be rejected per se. If the prompt contains sufficient individual and creative information about the type and appearance of the AI output, which is also objectively manifested in the output, this may justify protection under copyright law as a work. The author would then be the prompt writer.

Written by

Marco Erler

LAUSEN

TIER 1

Bird & Bird

DLA Piper

FREY Rechtsanwälte >

Irle Moser

LAUSEN >

Raue

Redeker Sellner Dahs

SKW Schwarz