Japan’s Content Overseas Distribution Association (CODA), which represents the nation’s most prominent intellectual property holders, including Studio Ghibli, Bandai Namco, and Toei Animation, has demanded that OpenAI stop using its members’ copyrighted material to train its video generation model, Sora 2.
The letter, first reported by Automaton, accuses OpenAI of using Japanese creative works without authorization, calling the act of replication during the machine learning process a potential “copyright infringement.” CODA argued that the company’s generative AI system, which has produced numerous clips and visuals bearing a strong resemblance to Japanese animation and gaming characters, has crossed legal lines.
“CODA considers that the act of replication during the machine learning process may constitute copyright infringement,” the organization said.
Register for Tekedia Mini-MBA edition 19 (Feb 9 – May 2, 2026): big discounts for early bird.
Tekedia AI in Business Masterclass opens registrations.
Join Tekedia Capital Syndicate and co-invest in great global startups.
Register for Tekedia AI Lab: From Technical Design to Deployment (next edition begins Jan 24 2026).
The complaint follows the September 30 launch of Sora 2, which quickly became a global sensation for its lifelike AI-generated videos — many of which adopted the visual language and style of beloved Japanese animation studios. In response, Japan’s Ministry of Economy, Trade, and Industry reportedly asked OpenAI to stop replicating Japanese artwork, underlining the rising alarm among Japanese creators about generative AI’s impact on cultural property.
This controversy marks one of several global confrontations between AI developers and copyright owners. Over the past year, a wave of “cease and desist” letters has been issued to major AI companies, including OpenAI, Stability AI, and Midjourney, over their alleged use of copyrighted material without permission. Many of these warnings have escalated into full-scale lawsuits.
In the United States, The New York Times filed a landmark lawsuit against OpenAI and Microsoft in December 2023, accusing them of using millions of its articles to train GPT models without authorization. The Times alleged that the AI systems could reproduce entire paragraphs from its reporting, potentially replacing its journalism in search results and chat interfaces. Similarly, several prominent authors — including George R.R. Martin, John Grisham, and Jodi Picoult — sued OpenAI, claiming the company’s large language models were trained on pirated copies of their books. Getty Images also sued Stability AI in both the U.S. and the U.K., accusing it of using more than 12 million copyrighted photos to train its image generator, Stable Diffusion.
The rising volume of legal challenges has forced AI companies to rethink their approach to data sourcing. In an effort to secure lawful access to training data, OpenAI has begun signing content partnerships with major platforms. Earlier this year, it struck deals with Reddit, Stack Overflow, the Associated Press, and Axel Springer, the parent company of Business Insider and Politico, to license their data for model training. These partnerships mark a shift in strategy as AI firms face increasing scrutiny from regulators and creators worldwide.
However, CODA insists that OpenAI’s reliance on “opt-out” systems — where content owners must explicitly request exclusion from training — violates Japan’s copyright regime.
“Under Japan’s copyright system, prior permission is generally required for the use of copyrighted works, and there is no system allowing one to avoid liability for infringement through subsequent objections,” the association said in its statement.
The group urged OpenAI to “respond sincerely” to its members’ concerns and immediately cease using Japanese intellectual property without prior authorization. The association also hinted that it could pursue legal or diplomatic action if the company fails to comply.
The growing number of copyright disputes highlights a widening fault line between AI innovation and intellectual property protection. As AI models become more advanced and capable of mimicking human artistry, creative industries across the world — from Hollywood to Tokyo — are mobilizing to safeguard their work from unauthorized digital reproduction.
It is believed that Japan’s pushback could become a defining test of how nations with robust copyright protections adapt to the AI age. With AI-generated content blurring the lines between inspiration and imitation, OpenAI and its rivals face an increasingly urgent challenge: balancing technological progress with the rights of the creators who supply the culture that fuels it.