Midjourney Demands Studios Suing It Disclose Their Own AI Data

Midjourney in July 2026 filed a motion in U.S. federal court, requiring film studios-plaintiffs to disclose how they themselves apply AI within their companies. Instead of a traditional passive defense, the defendant took an active stance: if you accuse us of copyright infringement through AI training, prove that you don't do the same. This was reported by Variety.

What are the studios accusing Midjourney of?

The studios accuse the AI image generator of training its model on copyrighted characters without permission and without compensation. Such claims became standard in the AI-copyright domain since 2023: copyright holders insist that training neural networks on others' works represents unauthorized reproduction and a violation of exclusive rights.

Midjourney is not the first company in such a position. In different periods, similar lawsuits were filed against OpenAI, Stability AI, and Google. Such disputes touch on a fundamental legal question: is training an AI model on publicly available images "fair use" under American law or copyright infringement? American courts have yet to develop a unified answer.

Midjourney's counterattack is unconventional: the company does not limit itself to denying the accusations, but itself goes on the offensive, demanding that the plaintiff studios disclose internal data about their own use of AI.

• Studios accuse Midjourney of training the model on rights holders' characters without permission
• Similar lawsuits in 2023–2025 were filed against OpenAI, Stability AI, and Google
• Midjourney filed a motion with a U.S. federal judge demanding compulsory data disclosure
• Without court intervention, studios have the right to refuse to provide such information
• Information about the motion was published in Variety in July 2026

Why does

Midjourney demand data from the plaintiffs themselves?

The logic of the defensive strategy is based on the principle of symmetric accountability. Major film studios are actively implementing AI in production — from storyboard generation and concept visualization to post-production, voice acting, and special effects processing. If it turns out that in doing so they use AI tools trained on others' materials without proper licenses, their position in court will be extremely vulnerable.

This is precisely what Midjourney is counting on. The company is essentially asking the court a question: before holding us accountable for AI training practices, determine whether the plaintiffs themselves adhere to the standards they demand from us. If the studios' AI practices prove similar to those being challenged, this could shift the course of the proceedings — opening the way either to a settlement or to a discussion about what constitutes "industry standard" when forming training datasets.

In American procedural law, the mechanism for such disclosure is called discovery — a party has the right to demand that an opponent provide documents if they may be relevant to the case. This is precisely the tool Midjourney has employed.

The strategy works to the defendant's advantage in any scenario: studios either prove the legal integrity of their AI practices — and Midjourney gains a clear understanding of the opponent's arguments — or discover their own vulnerabilities, which could change the balance of power long before a ruling is issued.

What this means

Midjourney's motion is an attempt to establish a new standard in AI copyright disputes: to achieve transparency not only from the defendant, but also from the plaintiff. If the federal court grants it, this will create a precedent for the entire generative AI industry. Copyright holders planning similar lawsuits will be forced to account for the fact that their own AI practices may become the subject of judicial scrutiny — and could work against them.