A coalition of major Japanese creative studios, including the renowned Studio Ghibli and video game publisher Bandai Namco, has formally accused OpenAI of copyright infringement. The dispute centres on the training data used for OpenAI’s advanced AI video model, Sora 2.
The Content Overseas Distribution Association (CODA), an anti-piracy organization representing these intellectual property holders, sent a formal letter to OpenAI last week. The letter demands that the company stop using its members’ copyrighted content to train the Sora 2 model.
In its communication, CODA stated it “considers that the act of replication during the machine learning process may constitute copyright infringement.” This legal challenge follows observations that Sora 2 has generated content featuring copyrighted characters owned by the represented studios.
Studio Ghibli, Bandai Namco, Square Enix demand OpenAI stop using their content to train AI https://t.co/qi834KfYPn
— The Verge (@verge) November 3, 2025The issue has attracted the attention of the Japanese government, which has also formally requested that OpenAI stop replicating Japanese artwork in its AI systems.
The complaint involves some of Japan’s most iconic creative exports. Studio Ghibli, famous for its hand-drawn animated films, has seen its distinctive art style widely mimicked by AI image generators since the launch of ChatGPT 4o.
Read: OpenAI Signs $38 Billion Cloud Deal with Amazon AWS for AI Compute
Bandai Namco, the publisher known for global hits like “Elden Ring” and “Tekken,” has experienced the reproduction of its characters and game content by the Sora 2 model. This formal action represents a significant escalation in the ongoing global debate regarding the use of copyrighted material for training generative AI systems.
