![[EXCLUSIVE] Dragon’s Dogma 2: Dark Arisen Team Opens Up About the Expansion’s Name, New Norgan Region, and Performance Goals](https://cdn.gamerbraves.com/2026/07/Dragons-Dogma-2-Dark-Arisen_Interview_FI-360x180.jpg)
A group representing Japan’s largest entertainment companies has asked OpenAI to stop using their content to train the Sora 2 video AI without permission. The Content Overseas Distribution Association (CODA) submitted a formal request to OpenAI CODA represents major anime, game, and manga publishers in Japan, including Aniplex, Bandai Namco, Studio Ghibli, Square Enix, Kadokawa, and Shueisha.
The organization said it took action after member companies raised concerns about how Sora 2 was trained. According to CODA’s announcement, they found that many of Sora 2’s video outputs look very similar to Japanese content and images, including specific copyrighted works. CODA believes this shows that Japanese content was used as training data for the AI model.
CODA warned that the copying process involved in machine learning “may constitute copyright infringement.”
While OpenAI has set up an opt-out system that lets copyright holders request their content be excluded, CODA argues this approach doesn’t align with Japanese copyright law. Under Japan’s legal framework, CODA states that permission must be obtained before using copyrighted works. The organization emphasized that “there is no system allowing one to avoid liability for infringement through subsequent objections.”
In their request, CODA asked OpenAI to stop training on content without authorization. They also called on the company to “respond sincerely to claims and inquiries from CODA member companies regarding copyright infringement related to Sora 2’s outputs.”
