AI Art vs. The Artist
Apr. 11, 2025 • Kamisetty Harika
Ghibli art holds a special place in the hearts of myriads of people. It is a legendary Japanese animation style, associated with Studio Ghibli, founded by Hayao Miyazaki, Isao Takahata, and Toshio Suzuki.[1] The art/style has inspired generations with its intricate visuals and emotional storytelling approach. Recently, AI-generated imagery mimicking the style of the production company Ghibli studio began flooding the social media platforms, credits to the new filter available in OpenAI’s GPT-4o. While some are beguiled by the aesthetics of the Ghibli world, devoted fans and artists are outraged and expressed their disappointment on how the AI-generation imagery undermined the human touch, artistic legacy of Studio Ghibli and its art.[2]
As the re-imaging took storm and the debate gained momentum, an old video of the co-founder, Miyazaki resurfaced, where he lashed out and strongly condemned AI-generated art, which could replicate the Ghibli-style. In the clip, he passionately stated, “I am utterly disgusted. If you really want to make creepy stuff, you can go ahead and do it. I would never wish to incorporate this technology into my work at all. I strongly feel this is an insult to life itself.”[3]
The Controversy and Legal Questions
Meanwhile, various lawsuits against OpenAI have been filed from artists across the world, against OpenAI and several other companies, arguing copyright infringement. Sam Altman, OpenAI’s CEO, responded acknowledging the roaring demand for AI-generated Ghibli-style images, even stating that the company’s “GPUs are melting” due to the tool’s popularity.[4] OpenAI has since been forced to limit the access to the feature and its usage. When asked by news agency about potential infringement of Studio Ghibli’s intellectual property, OpenAI responded that it was still improving its model, intending to give its users as much creative freedom as possible.[5]
This situation raises certain important questions. Is the Ghibli style protected under copyright law? If so, is ChatGPT violating its copyright by generating images in their style? Has ChatGPT been trained using Studio Ghibli’s previous artworks, and if so, would it constitute fair use under the Copyright Law? Reports suggest that OpenAI is aggressively lobbying the U.S. Congress and the White House to expand fair use laws to accommodate AI-generated content. Currently, fair use applies to search engines, satire, and memes, allowing copyrighted material to be used without direct permission. OpenAI claims its AI model follows a “conservative approach” to imitating individual artists’ styles, to avoid direct replication of artists’ works and legal trouble.[6]
Understanding Copyright and Fair Use
Copyright relates to literary and artistic creations, such as books, music, paintings and sculptures, films and technology-based works (such as computer programs and electronic databases). The right of copyright owners to prevent others from making copies of their works without permission is the most basic right protected by copyright legislation. The ideas in the work do not need to be original, but the form of expression must be an original creation by the author.
The laws of many countries recognize the concepts of fair use or fair dealing. There are several types of limitations and exceptions to copyright protection. The two basic types of limitations and exceptions in this category are: i) Fair use (or fair dealing) which allows the use of the work without permission for the purposes of criticism, commentary, teaching, research, parody, satire and etc. It carries no obligation to compensate the right owner. ii) Non-voluntary (compulsory) licenses allow restricted use of the copyrighted works and requires compensation or royalty to be paid to the creator in case of any breach or non-authorised use. The Marrakesh Treaty provides for the adaptation of these limitations and exceptions for creation and cross-border transfer of published works. This ensures that the rights of creators are protected while keeping works available to the public. [7]
Is AI-Generated Ghibli Art Infringing Copyright?
Applying the legal principle of Res Ipsa Loquitur (meaning “the thing speaks for itself”), it may be argued that the AI’s outputs are influenced by Ghibli’s artistic style, making it evident that AI models, including ChatGPT, have been trained using the previous artworks of Studio Ghibli and its artists. However, the demand for AI-generated Ghibli-style art is tremendous and cannot be ignored. People are captivated by their Ghibli-version and are loving the magic of the Ghibli-like world. The appeal is undeniable. And with this level of interest, is an outright ban really the answer?
A Middle-Ground Solution: Licensing and Royalty Agreements
OpenAI’s method of creating images differs from traditional hand-drawn animation, using code and GPU processing rather than manual brush strokes, its process (OpenAI’s method) may be copyrightable in its own right, while the style of the original Ghibli belongs to its original creators or artists. Therefore, I think, a practical solution to this issue, could be licensing. Instead of restricting AI-generated Ghibli art, Studio Ghibli, along with its artists and OpenAI could establish licensing agreements, including the rates of royalty, thereby, fairly compensating and crediting the creators, whilst allowing the public to enjoy AI-generated re-imaginings.
Moreover, the legacy of Studio Ghibli and its creations are well-known, thereby causing no confusion about the authorship and the originality of its artistic style. Further, if Miyazaki or Studio Ghibli decide to take legal action, negotiating licensing agreements would, in my opinion, be the most reasonable solution.
This debate is not just about AI and copyright but about passion, nostalgia, and the love people have for Ghibli’s artistry. While the AI tools have exciting potential, they also challenge complex ethical and legal issues. The answer, in my opinion, does not necessarily lay in bans and lawsuits, but in fair compensation and collaboration. By acknowledging and honouring the original creators while adapting to the evolving digital world, we can discover a middle ground that serves both creators and fans of this timeless art form.
References:
The author affirms that this article is an entirely original work, never before submitted for publication at any journal, blog or other publication avenue. Any unintentional resemblance to previously published material is purely coincidental. This article is intended solely for academic and scholarly discussion. The author takes personal responsibility for any potential infringement of intellectual property rights belonging to any individuals, organizations, governments, or institutions.