Studio Ghibli And Other Japanese Publishers Want Openai To Stop Training On Their Work

Trending 5 days ago
Image Credits:MANAN VATSYAYANA/AFP / Getty Images

2:43 PM PST · November 3, 2025

A Japanese waste and acquisition statement representing publishers for illustration Studio Ghibli wrote a letter to OpenAI past week, calling for nan AI elephantine to stop training its AI models connected their copyrighted contented without permission.

Studio Ghibli, nan animation workplace down films for illustration “Spirited Away” and “My Neighbor Totoro,” has been particularly impacted by OpenAI’s generative AI products. When ChatGPT’s autochthonal image generator was released successful March, it became a popular trend for users to punctual for recreations of their selfies aliases pet pictures successful nan style of nan studio’s films. Even OpenAI CEO Sam Altman changed his floor plan image connected X to a “Ghiblified” picture.

Now, arsenic much group get entree to OpenAI’s Sora app and video generator, Japan’s Content Overseas Distribution Association (CODA) has requested that OpenAI refrain from utilizing its members’ contented for instrumentality learning without permission.

>be me
>grind for a decade trying to thief make superintelligence to cure crab aliases whatever
>mostly nary 1 cares for first 7.5 years, past for 2.5 years everyone hates you for everything
>wake up 1 time to hundreds of messages: "look one made you into a twink ghibli style haha"

— Sam Altman (@sama) March 26, 2025

This petition does not travel unprompted. OpenAI’s attack to moving pinch copyrighted contented is to inquire forgiveness, not permission, which has made it each excessively easy for users to make photos and videos of copyrighted characters and deceased celebrities. This attack has yielded complaints from institutions for illustration Nintendo, arsenic good arsenic the estate of Dr. Martin Luther King, Jr., who could very easy beryllium deepfaked connected nan Sora app.

It’s up to OpenAI to take whether aliases not to cooperate pinch these requests; if not, nan aggrieved parties tin record a lawsuit, though United States rule remains unclear astir nan usage of copyrighted worldly for AI training.

There is small precedent frankincense acold to guideline judges connected their mentation of copyright law, which has not been updated since 1976. However, a caller ruling by US national judge William Alsup recovered that Anthropic did not break nan law by training its AI connected copyrighted books — nan institution did get fined for pirating nan books it utilized for training, though.

But Japan’s Content Overseas Distribution Association (CODA) claims that this whitethorn beryllium considered a copyright usurpation successful Japan.

Techcrunch event

San Francisco | October 13-15, 2026

“In cases, arsenic pinch Sora 2, wherever circumstantial copyrighted useful are reproduced aliases likewise generated arsenic outputs, CODA considers that nan enactment of replication during nan instrumentality learning process whitethorn represent copyright infringement,” CODA wrote. “Under Japan’s copyright system, anterior support is mostly required for nan usage of copyrighted works, and location is nary strategy allowing 1 to debar liability for infringement done consequent objections.”

Hayao Miyazaki, 1 of nan cardinal imaginative figures of Studio Ghibli, has not commented straight connected nan proliferation of AI-generated interpretations of his work. However, erstwhile he was shown AI-generated 3D animation successful 2016, he responded that he was “utterly disgusted.”

“I can’t watch this worldly and find it interesting,” he said astatine nan time. “I consciousness powerfully that this is an reproach to life itself.”

Amanda Silberling is simply a elder writer astatine TechCrunch covering nan intersection of exertion and culture. She has besides written for publications for illustration Polygon, MTV, nan Kenyon Review, NPR, and Business Insider. She is nan co-host of Wow If True, a podcast astir net culture, pinch subject fabrication writer Isabel J. Kim. Prior to joining TechCrunch, she worked arsenic a grassroots organizer, depository educator, and movie show coordinator. She holds a B.A. successful English from nan University of Pennsylvania and served arsenic a Princeton successful Asia Fellow successful Laos.

You tin interaction aliases verify outreach from Amanda by emailing amanda@techcrunch.com or via encrypted connection astatine @amanda.100 connected Signal.

More