Midjourney is pushing a federal judge to compel Disney, Universal, and Warner Bros. Discovery to disclose how they use artificial intelligence internally as part of a copyright dispute.

The case centers on whether Midjourney’s AI-generated images infringe copyrighted works and how a studio’s training data and internal AI practices should factor into the fight.

The suit began in June 2025 when Disney and Universal accused Midjourney of enabling broad copyright infringement, calling Midjourney a “bottomless pit of plagiarism” for recreations like Darth Vader and Elsa.

Representatives for the studios argue the goal is to stop unauthorized copying and derivative works, not to expose internal AI methods.

Midjourney’s motion seeks to counter-sue by forcing transparency about studios’ AI usage, including potential disclosure of their training practices.

Midjourney contends studios may be using internal, non-public AI processes for storyboard ideation and development, which could bolster Midjourney’s defenses.

The studios’ lead attorney labeled Midjourney’s request a fishing expedition, insisting the aim is to curb infringement without hindering AI development generally.

Disney has signaled strong AI interest, including a reported $1 billion investment in OpenAI to bring Disney characters to a platform, though plans shifted with SoraAI's developments.

A judge previously limited disclosures to consumer-facing AI usage; Midjourney seeks to overturn this to prevent cherry-picking evidence.

The motion argues that access to internal AI usage data could be relevant to Midjourney’s fair use and unclean hands defenses.

Midjourney maintains that training on copyrighted characters qualifies as fair use and seeks broader production of internal AI usage documentation beyond consumer outputs.

SoraAI was discontinued earlier, but Disney says it will continue engaging with AI platforms in ways that respect IP and creator rights, influenced by ongoing litigation and potential disclosures.