Japanese creators ask OpenAI to respect copyright following launch of Sora 2

Japan's Content Overseas Distribution Association (CODA) is asking OpenAI to respect the copyright of creators in the operation of its video-audio generation system, Sora 2.

In a statement posted on its website, CODA said it has submitted a written request to the artificial intelligence company at the behest of its member companies.

They ask that their content would not be used for machine learning without permission and that OpenAI would respond to claims and inquiries about copyright infringement over Sora 2's outputs.

"CODA has confirmed that a large portion of content produced by Sora 2 closely resembles Japanese content or images. CODA has determined that this is the result of using Japanese content as machine learning data," the statement read.

"In cases, as with Sora 2, where specific copyrighted works are reproduced or similarly generated as outputs, CODA considers that the act of replication during the machine learning process may constitute copyright infringement."

According to CODA, OpenAI's opt-out system, where copyright holders can request exclusion of their works from AI training, does not align with Japan's copyright system where prior permission is required before use.

CODA has 36 corporate members as of April 2025, including Studio Ghibli, Toei Company, Aniplex, Shueisha, Shogakukan, Square Enix, and more animation, manga, broadcast, and publication companies. It also has 10 supporting members, including Sony Music Japan and Nippon Animation.

Meanwhile, OpenAI is the company behind ChatGPT. It launched Sora 2 on September 30 as the successor to Sora, its text-to-video model made available in December 2024 for ChatGPT Plus and Pro users.

Along with the new model, a Sora app has been released and is currently supported in the United States, Canada, Korea, Japan, Vietnam, Thailand, and Taiwan. It features a social media video feed, where users can make a "cameo" on generated scenes. —MGP, GMA Integrated News