Amid legal battles with Disney, Universal, and Warner Bros over alleged copyright infringement, AI image generator Midjourney is pushing a US court to compel studios to reveal their internal use of AI technology. The startup argues that studios’ own AI practices could support Midjourney’s defense and reshape the scope of discovery.
- Midjourney demands studios reveal all internal AI usage details.
- Studios argue against broad disclosure, calling it a fishing expedition.
- The case could set a precedent on AI training and fair use defenses.
What happened
Midjourney filed a court motion to compel Disney, Universal, and Warner Bros. to provide comprehensive records of how they use artificial intelligence internally. This move is part of the companies' ongoing copyright litigation accusing Midjourney of training AI on protected studio content and generating images of copyrighted characters without authorization. While studios have demanded to halt Midjourney’s unauthorized use of their intellectual property, Midjourney denies wrongdoing and insists its actions are protected by the fair use doctrine.
The judges had previously ordered studios to share information related only to AI systems that create public-facing images and videos. Midjourney challenges this restriction, asserting it unfairly limits discovery to only evidence supporting the studios' claims, while excluding internal AI practices that might bolster Midjourney’s defenses. The company also requests disclosure of all internal prompts and outputs generated via its AI system by studio personnel, arguing these are critical to understanding the scope of alleged infringement.
Why it matters
The case illustrates the complex challenges in defining fair use in AI training, especially as studios themselves increasingly employ generative AI for internal creative work such as storyboarding and idea development. Midjourney’s request highlights potential industry-wide practices where studios might also use unlicensed copyrighted material to train their AI, which could undermine their accusations against Midjourney.
If courts broaden the discovery scope to include internal AI usage, it may set a significant precedent influencing how copyright law applies to AI-generated content and the responsibilities of both creators and corporations using protected works in training datasets. This could affect not only Midjourney but broader AI development and content creation industries, impacting future regulatory and legal standards around intellectual property and AI.
What to watch next
Observers should monitor how the court rules on Midjourney’s discovery motion as this will signal whether internal AI use by studios must be disclosed in copyright litigation. The decision could shape the evidentiary standards for AI-related copyright cases by clarifying if internal AI training practices have a role in fair use defenses. Additionally, the outcome may influence how studios and AI companies negotiate the boundaries of creative content use and AI innovation moving forward.
Industry stakeholders will also watch for any further legal arguments regarding the distinction between consumer-facing AI outputs and internal AI uses, as well as how courts treat demands for AI prompts and generated outputs. These developments could define the transparency requirements studios and AI tool operators must meet when their technologies intersect with copyrighted materials.