Midjourney wants Hollywood studios to reveal the details of their AI usage

What has Midjourney actually asked the court to do?

Midjourney has filed a motion asking Judge John Kronstadt to force Disney, Universal and Warner Bros to hand over internal details of their own AI use, including training datasets, model weights and board level presentations on generative AI. A magistrate judge had previously limited discovery to only "consumer facing" AI tools, and Midjourney is pushing back on that limit.

The underlying case started in June 2025, when Disney and Universal sued Midjourney over AI generated images of characters such as Bart Simpson and Darth Vader. Warner Bros Discovery joined in September 2025, adding Batman and Superman to the complaint. Midjourney's counter argument is essentially "unclean hands": if the studios use similar unlicensed AI methods internally for storyboarding or ideation, that undermines their claim that Midjourney's approach is uniquely unlawful.

Why should a small business owner in the UK care about a Hollywood copyright case?

Because it is another sign that AI generated content and copyright law are nowhere near settled, and every small business using AI tools for images, marketing or branding is operating in that same grey area, just with far less legal cover than a studio.

This isn't an abstract legal squabble. If a US federal judge decides studios must disclose their own AI training and usage practices, it will shape how courts everywhere, including the UK, think about who owns AI generated content and what "fair use" of copyrighted material for AI training actually means. Small businesses commissioning AI generated logos, product photos or ad creative are downstream of all of this.

Does this affect UK businesses directly, or is it just a US legal matter?

It affects UK businesses indirectly but meaningfully, because UK courts are wrestling with the same underlying question and have not yet given a clear answer. On 4 November 2025, the UK High Court ruled on Getty Images v Stability AI, rejecting most of Getty's copyright claims but finding a narrower trademark infringement over reproduced watermarks. UK lawyers have been clear that this leaves AI image copyright law still largely unsettled here too.

Combine that with a legal principle that already applies in the UK: AI generated images with no meaningful human creative input are not protected by copyright. If you generate a logo or hero image purely from a text prompt with no editing or creative direction, you may not actually own the copyright to it, and neither does anyone else claiming to license it to you.

What should I actually do differently because of this?

Treat any AI generated image used in your branding, website or ads as a commercial risk until you've checked how it was made and what licence you're getting. Three concrete steps:

• Favour tools trained on licensed data for anything customer facing or higher stakes, such as Adobe Firefly, Shutterstock AI or Getty AI, rather than general purpose image generators.
• Keep a record of the prompts, edits and creative decisions you made when producing any AI image you plan to use commercially, since meaningful human input is what makes it copyrightable in the UK.
• Read the licence terms of whatever tool you use before you build a logo, product shot or ad campaign around its output, not after.

Is AI adoption actually worth the risk for a small business?

Yes, the risk is manageable if you're deliberate about where and how you use AI. UK SME adoption of AI hit 54% in 2026, up from 35% in 2025 and 25% in 2024, according to the British Chambers of Commerce. A QuickBooks survey from January 2026 found 70% of UK SMEs use AI regularly, mostly through built in features in accounting, CRM and e-commerce tools rather than standalone AI image generators.

That distinction matters. Most of the safe, high value AI use for small businesses isn't image generation at all. It's automating bookings, following up with customers who never rebooked, and making sure your business actually shows up when someone asks Google or an AI assistant to recommend a local plumber, salon or restaurant.

What's Braynex Services' take on this?

Our view is that this case is a useful reminder, not a reason to panic. Most small businesses we work with aren't at risk from Midjourney style copyright disputes because they aren't building brand assets from raw AI image prompts in the first place. The bigger risk we see day to day is businesses renting their entire online presence, a Facebook page, a Linktree, a Fresha or Booksy listing, and having no control over the platform, the data or the terms.

We saw this directly with a nail salon paying Fresha around £1,800 a month in commission before moving to its own booking system at £35 a month, saving roughly £21,000 a year. The lesson is the same one this court case points to: ownership beats renting, whether that's your booking system or your brand assets. If you don't own the underlying asset, whether it's a customer list or an image's copyright, you don't control what happens to it.

The practical opportunity right now isn't legal risk avoidance, it's visibility. Being recommended by AI assistants like ChatGPT, Gemini and Google AI Overviews is becoming the new local SEO, and that depends on having a real website and a verified Google Business Profile, not a rented platform page.

If you want a clear, no obligation look at where your business is exposed, whether that's rented platforms, missed calls, or an outdated Google presence, book a free audit at braynexservices.com.