IA · 5 July 2026 · 3 min read
Midjourney's Countermove: Demanding Hollywood Reveal Its Internal AI Usage
In brief: The legal battle between Midjourney and major Hollywood studios is escalating. Accused of copyright infringement, the startup is now seeking to compel giants like Disney, Universal, and Warner Bros. to disclose their internal generative AI practices, aiming to expose potential double standards between public lawsuits and behind-the-scenes production workflows.
by Team Mocchi's
The Tug-of-War Between Training and Copyright
The tension between entertainment giants and generative artificial intelligence startups has reached a critical turning point. At the heart of the conflict is a high-profile lawsuit initiated by major film studios, including Disney, Universal, and Warner Bros., accusing a leading image-generation platform of copyright infringement. The plaintiffs claim that the AI models were trained on protected visual assets without authorization, enabling users to generate highly detailed replicas of iconic characters ranging from Darth Vader to Bart Simpson.
While the tech startup defends the legitimacy of its training methods under the principle of fair use—arguing that it constitutes a transformative and creative evolution—the legal battle has shifted to an unexpected tactical ground. Rather than sticking solely to technical arguments about algorithms, the company has launched a counter-offensive focused on corporate transparency, demanding that the studios disclose their own practices.
The Discovery Dispute and the Issue of Internal Use
The current clash centers on the discovery process, where both parties must share relevant documents before going to trial. Previously, a federal judge ruled that the movie studios had to provide information regarding their use of generative AI, but limited this obligation to public-facing materials, such as promotional images or commercially distributed videos.
The startup has filed a new motion challenging this limitation, calling it an unfair filter that allows the studios to cherry-pick only the documents that support their claims of financial harm, while hiding potentially contradictory internal workflows. The defense aims to prove that, behind closed doors, major studios are already heavily relying on the very technology they are fighting in court.
The Double Standard in the Creative Industry
The core of the defense touches on a highly sensitive issue for the entertainment world. If major studios are utilizing generative AI for internal operations—such as creating storyboards, pre-visualizing scenes, or brainstorming concepts for movies and television shows—while training these models on unlicensed materials, their legal stance against AI startups would be severely undermined.
Demonstrating that internal AI usage is a standard industry practice, even among the entities suing for infringement, would strengthen the argument that training models on existing data represents a natural technological progression rather than a malicious effort to disrupt the market. The startup hopes to expose this industrial paradox to counter the massive damage claims filed by the studios.
Prompt History Under Scrutiny
Another key element of the motion concerns how the copyright infringement evidence was generated. Legal teams representing the studios presented highly accurate images of copyrighted characters produced by the platform. In response, the defense is demanding access to the full history of text prompts and outputs generated by the studios during their testing.
The startup suspects that these infringing results were not the product of standard user behavior, but rather the result of highly specific, repeated, and targeted instructions designed to force the algorithm into reproducing copyrighted material. Accessing the complete generation logs would help prove that the copyright violations were artificially engineered for the sake of the lawsuit rather than being a natural byproduct of regular tool usage.
Future Implications for Software Development and Intellectual Property
The outcome of this legal battle will reverberate far beyond Hollywood, impacting any enterprise developing or integrating artificial intelligence solutions. Should the court grant the startup's request, the resulting transparency mandate could force major corporations to re-evaluate their confidentiality agreements and internal AI deployment policies.
For software creators and AI integrators, the ruling will help define the operational boundaries of model training and the practical scope of the fair use defense. A favorable decision for the startup could accelerate the integration of generative systems in industrial design and pre-production workflows, legitimizing practices that currently exist in a regulatory gray area. Conversely, stricter disclosure rules could impose heavy compliance burdens, slowing down the adoption of custom enterprise AI.