Hollywood vs. The "Bottomless Pit of Plagiarism"
Three of Hollywood's "Big Five" studios have now sued Midjourney, with damages that could exceed $20 million. Disney, Universal, and Warner Bros. describe the AI image generator as a "bottomless pit of plagiarism" and a "virtual vending machine" for unauthorized copies of their copyrighted characters. The lawsuits threaten the entire foundation of AI-generated art.
The complaints aren't abstract legal theory—they're illustrated with page after page of AI-generated images showing Batman, Mickey Mouse, Bugs Bunny, and hundreds of other copyrighted characters produced from simple text prompts. Only Sony and Paramount remain on the sidelines. The legal reckoning for generative AI has arrived.
The Disney-Universal Lawsuit (June 2025)
The first major blow came in June 2025, as reported by CNN:
$150,000 Per Infringement: The studios seek statutory damages for each infringed work, with over 150 works listed in the complaint
Screenshot Evidence: The complaint functions as "a gallery exhibition of infringement"—pages of AI-generated images clearly replicating copyrighted characters
Simple Prompts: Characters could be generated with one-word prompts like "screencap," producing detailed images of Luke Skywalker, Iron Man, or the Joker
Injunctive Relief: Beyond damages, the studios want court orders preventing future infringement
"The studios describe Midjourney as a 'bottomless pit of plagiarism' and a 'virtual vending machine' for unauthorized copies of their works—and they've brought the screenshots to prove it." — Copyright Lately
Warner Bros. Escalation (September 2025)
Three months later, Warner Bros. joined the fight with an 87-page complaint:
Character Evidence: AI-generated images of Superman, Batman, Wonder Woman, Bugs Bunny, Scooby-Doo, Tom and Jerry, Powerpuff Girls, and Rick and Morty
"Systematic and Willful": According to IPWatchdog, WBD alleges the infringement isn't accidental but deliberate and ongoing
Ignored Cease-and-Desist: Multiple requests to stop infringement went unheeded
Video Expansion: The complaint notes Midjourney has launched video generation services, potentially expanding infringement into new territory
The Technical Evidence
The lawsuits cite technical research demonstrating the problem:
IEEE Spectrum Study (January 2024): Researchers showed Midjourney V6 could generate near-replicas of copyrighted characters even without explicitly naming them
One-Word Replication: Single-word prompts produced detailed, recognizable versions of protected characters
No Meaningful Guardrails: Despite multiple warnings, Midjourney hasn't implemented safeguards to prevent copyrighted content generation
Midjourney's Defense
In an August 2025 response documented by Artnet, Midjourney's attorneys argued:
Fair Use: Training AI models falls within allowed fair use parameters, similar to how humans learn by studying existing art
Studios Use AI Too: The plaintiffs themselves "use and profit from their use of Midjourney and other generative A.I. tools"
Learning Analogy: AI training is like humans "learning to draw or paint, not by memorizing individual artworks, but by internalizing patterns and techniques through repeated exposure"
"Midjourney's attorneys likened training AI models to how humans learn to draw or paint—'not by memorizing individual artworks, but by internalizing patterns and techniques through repeated exposure and practice.'" — Artnet
Why This Case Is Different
Copyright Lately analysis explains why previous AI copyright cases may not predict this outcome:
Output Focus: Previous cases focused on training data; these focus on infringing outputs that users can actually generate
Clear Evidence: The character reproductions are obvious and undeniable, not edge cases requiring technical analysis
Well-Funded Plaintiffs: Disney, Universal, and Warner Bros. have essentially unlimited legal budgets
Precedent Stakes: A ruling against Midjourney could reshape the entire AI image generation industry
The Streaming Threat
Midjourney's expansion makes the studios' concerns more urgent:
24/7 Streaming Channels: Midjourney now operates continuous streaming content on its website and YouTube
Television Market Entry: Discussion of adding "channels" suggests intent to compete directly with traditional media
Video Generation: Beyond images, Midjourney's video capabilities create new infringement vectors
Industry Implications
The lawsuits' outcome will shape AI art's future:
Licensing Models: A loss could force AI companies to negotiate licensing deals with content owners
Training Data Scrutiny: All AI image generators may face pressure to prove clean training data provenance
User Liability: Questions remain about whether users generating copyrighted content face personal liability
Chilling Effect: Even without final rulings, the litigation may slow AI art adoption in commercial contexts
The Bottom Line
Midjourney built a billion-dollar business on a legal foundation that's now crumbling. Three major studios, potential damages exceeding $20 million, and clear evidence of copyrighted character reproduction create an existential threat. The fair use defense may ultimately succeed, but the litigation will be expensive, lengthy, and damaging regardless of outcome. The age of consequence-free AI art generation is ending.
AI NEWS DELIVERED DAILY
Join 50,000+ AI professionals staying ahead of the curve
Get breaking AI news, model releases, and expert analysis delivered to your inbox.




