Millions of people are currently becoming enthusiastic about Chatgpt and the ability to create pictures in the Ghibli style.
If you are currently looking at X.com or other social media, you will find that you can find many pictures that are very similar to Studio Ghibli’s style. This is because Chatgpt-4O can create content in the style of a specific author.
However, the problem is that Chatgpt must have bought this advantage with hard copyright infringement: Chatgpt-4o must have been trained with content that are connected with the studio founded by Hayao Miyazaki, based on it or were generated by it. And AI like Chatgpt has long been suspected of copying material.
Students spy on passers -by with a hidden camera and a AI, find out the name and address in seconds
More videos
Autoplay
Chatgpt can create pictures in the style of Ghibli, but behind it is protected material
What is the problem? Material or films such as Studio Ghibli are protected by copyright. The rights on the films are not inherently and must therefore not simply be used by everyone. And that is a problem that Chatgpt and Openai has been plagued since it was founded.
Chatgpt was trained with an enormous amount of data from the Internet – from websites, books, articles on social media to academic articles. Contents that are freely accessible, but not for nothing. Not every picture that you can find on the Internet, you can use free of charge, except the author/photographer deliberately releases his work under the Creative Commons license.
However, it is also important that you cannot prevent anyone from creating works in the style of a person or a studio. Because the style itself is not protected. It’s like the discussion about fan art: you can create a picture of a Pokémon yourself and hang it on your room wall and nobody will be bothered by it. However, as soon as you start selling these works of art, it will be legally difficult.
AI companies generally rely on the “Fair Use” doctrine for their data, but is particularly valid in the USA. A US court recently decided that the training of AI with copyrighted content does not fall under the fair use doctrine (via Advomare.de). However, it is important: the judgment only affects non-generative AI and is therefore less relevant for AI models such as Chatgpt.
What does Openai about? Openaai claims that a ban has now been introduced to prevent images or works of art from being created that are supposed to create the style of a living artist. Dead artists, on the other hand, are okay. What is also okay is the copy of “studio styles”, which probably also includes Ghibli Studios (via XATAKA.com).
Complaints against Openaai and demands for protected material
How do the artists and authors see? Openaai and other AI companies have already been sued for hard violations of copyright law. This includes artists and image databases that defend themselves against the fact that Openai simply uses their content free of charge, even though they are not free of charge. Among other things, 17 large authors, including George RR Martin, have come together for a lawsuit against Chatgpt.
The accusation of the authors: the AI learns from texts that are illegally downloaded from the Internet. At the time, it was said in a statement that “generative AI threatens to destroy the profession of authors.”
Incidentally, Openaai is taking this as an opportunity to demand access to protected materials. Because if you don’t get access, China would win the AI race: Openaai says they urgently need access to legally protected material-because of China