Warner Bros. Discovery Sues Midjourney For Copyright Infringement
- Reference: 0179010716
- News link: https://yro.slashdot.org/story/25/09/04/2236226/warner-bros-discovery-sues-midjourney-for-copyright-infringement
- Source link:
> The company "brazenly dispenses Warner Bros. Discovery's intellectual property" by letting subscribers produce images and videos of iconic copyrighted characters, alleges the complaint, filed on Thursday in California federal court. "The heart of what we do is develop stories and characters to entertain our audiences, bringing to life the vision and passion of our creative partners," said a Warner Bros. Discovery spokesperson in a statement. "Midjourney is blatantly and purposefully infringing copyrighted works, and we filed this suit to protect our content, our partners, and our investments."
>
> For years, AI companies have been training their technology on data scraped across the internet without compensating creators. It's led to lawsuits from authors, record labels, news organizations, artists and studios, which contend that some AI tools erode demand for their content. Warner Bros. Discovery joins Disney and Universal, which earlier this year teamed up to sue Midjourney. By their thinking, the AI company is a free-rider plagiarizing their movies and TV shows. In the lawsuit, Warner Bros. Discovery points to Midjourney generating images of iconic copyrighted characters. At the forefront are heroes who're at the center of DC Studios' movies and TV shows, like Superman, Wonder Woman and The Joker; others are Looney Tunes, Tom and Jerry and Scooby-Doo characters who've become ubiquitous household names; more are Cartoon Network characters, including those from Rick and Morty, who've emerged as something of cultural touchstones in recent years. [...]
>
> The lawsuit argues Midjourney's ability to return copyrighted characters is a "clear draw for subscribers," diverting consumers away from purchasing Warner Bros. Discovery-approved posters, wall art and prints, among other products that must now compete against the service. [...] Warner Bros. Discovery seeks Midjourney's profits attributable to the alleged infringement or, alternatively, $150,000 per infringed work, which could leave the AI company on the hook for massive damages. The thrust of the studios' lawsuits will likely be decided by one question: Are AI companies covered by fair use, the legal doctrine in intellectual property law that allows creators to build upon copyrighted works without a license?
The lawsuit can be found [2]here .
[1] https://www.hollywoodreporter.com/business/business-news/warner-bros-discovery-sues-ai-company-copyright-infringement-1236361610/
[2] https://www.scribd.com/document/911515490/WBD-v-Midjourney-Complaint-Ex-a-FINAL-1#fullscreen&from_embed
"letting subscribers produce" (Score:2)
"letting subscribers produce images and videos of iconic copyrighted characters"
Midjourney is not the one committing the copyright infringement. Their users are (allegedly).
If Midjourney is encouraging users to commit copyright infringement, that would open them up to claims for contributory copyright infringement.
Midjourney may have a duty to take steps to prevent it, but it has yet to be established as to how far they should go in restricting users constitutionally protected freedom of expression in orde
Re: (Score:2)
Bullshit. If a user requests a copyrighted char, the user is putting in 0.1% or less, the rest is all Midjourney.
Re: (Score:1)
Their program clearly knows what Batman (etc.) looks like. It would be trivial for it to refuse to generate images that look like Batman.
Re: (Score:2)
The copying isn't in the generation. The copying happens when you make model weights from the input texts. At some level for some texts that might come under "fair use". Fair use, though, is an actual quantitative thing. Copy a whole text and you are in trouble. Copy one paragraph to comment on it and you'll be fine. Important texts are echoed enough in the training corpus, so whilst the places they are copied are probably fair use, likely the level of usage of their text in the LLM model weights will, in t
A garbage lawsuit. (Score:3)
Adobe makes Photoshop. Photoshop allows users to create images of Superman. The user can draw Superman, or put into a prompt, "Make this thing look like Superman". In both cases it is the *user* that is initiating the process to create Superman, which DC owns. Should Photoshop be banned?
The anti-AI trolls will pontificate that they are different, but they are not, not in a legal sense. The user was "trained" by watching Superman movies, so the user knows what superman looks like, and can thus draw him.
An LLM was "trained" by watching superman movies, so the LLM knows what superman looks like, and can thus draw him, *when prompted by the user.*
Courts have repeatedly found this to be the case. It's a garbage lawsuit. Throw it in the garbage.
Re: (Score:2)
If MidJourney and Photoshop are both tools, then so is a tool to download copyrighted films (which is clearly not respecting copyrights).
Copyright law already distinguishes between exact copies, derivative works, and fair use. All delineated by fuzzy boundaries. So it's contextual, based on circumstances. In the case of MidJourney, to comply with copyright law, they probably need to put up guardrails like GPT5 already has done. GPT5 will outright refuse to draw Superman, but MidJourney happily complies. If
Re: (Score:2)
And yet those tools are not illegal. Downloading a copyrighted file is considered not fair use. Making superman in Photoshop is, as long as it is for personal use. Likewise, using midjourney to make superman is the same, and should be considered fair use. If they want to sue the users who put these pictures on the web for others, then by all means they are free to do that, but suing Midjourney is just like suing Adobe. Garbage lawsuit.
Re: (Score:2)
In the real world where I live, we have both Xerox machines, computer printers, and this thing called The Internet. On my planet, nobody goes to the trouble of making one of those infringing images just to keep it framed in their living room for when their friends come over. No Siree, they do the work (however little or how much) and immediately PUBLISH and DISTRIBUTE that fucker. To at least one other person, but usually to as. may people as is possible. The Internet has a lot of possible in it.
The tools
Re: (Score:1)
You know what else photoshop does?
If you try to process a scanned image of currency, it will refuse to do so. It recognizes that you're working with currency imagery and refuses to comply.
[1]https://helpx.adobe.com/photos... [adobe.com]
[1] https://helpx.adobe.com/photoshop/cds.html
Re: (Score:2)
So the government does not hold copyrights on the image on a dollar bill. You can actually make a copy of it... but apparently there are regulations that your size has to be different (20% larger or smaller) and your printout can only be one sided. Photoshop probably does not have to legally prevent users from working with a scanned dollar bill, but believes that the pros of banning the use in this way outweighs the cons. But they don't *have" to, legally.
Re: (Score:1)
And yet they do.
Midjourney could also make such a decision.
Re: (Score:1)
Fun fact....and I guess the statute of limitations has expired, so I can say this.
When I was in 10th grade, I decided to do an experiment. I made photocopies of $5 bills. Black and white, 2 sided.
Then I went to a local govt building that had a change machine. It was the kind that had a tray that pushed in. You put the bill in the tray, pushed it in, and got change.
Well, let me tell you what. That old piece of shit was like a slot machine that couldn't lose. I went through my entire stack of $5 photocopied
Re: (Score:2)
> You know what else photoshop does?
> If you try to process a scanned image of currency, it will refuse to do so. It recognizes that you're working with currency imagery and refuses to comply.
> [1]https://helpx.adobe.com/photos... [adobe.com]
Yes, because there are only a few such images in the entire world, so it's easy to compare them. It would be simply impossible for the tool to check your upload against EVERY IMAGE IN THE UNIVERSE, which are all Copyright by default. Even a registry of a small number provided by a powerful group of moneyed stakeholders would be too hard to compare.
And if it did that, I would object. How does the tool know that I don't have rights to copy? How does it know I am going to publish? Get the fuck out of my busine
[1] https://helpx.adobe.com/photoshop/cds.html
Re: (Score:2)
Your argument is garbage. You sound like a disinformation AI. Companies aren't training one AI in the same sense one artist is trained. If an AI company wants 10,000 AI instances so it can support 10,000 simultaneous users, they aren't buying 10,000 copies of superman references to train off, which is how you would train 10,000 artists.
Re: (Score:2)
No, *your* argument is garbage, just like this lawsuit. The law makes no distinction between how an artist and an LLM is trained, because it is irrelevant. Courts have repeatedly ruled that this is irrelevant; what matters is what is produced and whether or not its fair use.
Did you know that I did not buy a single copy of a superman reference to train myself to be able to draw superman? Did you know that there's no law that says that somewhere in the past I had to have purchased *some* superman image to
Re: (Score:2)
> Adobe makes Photoshop. Photoshop allows users to create images of Superman. The user can draw Superman, or put into a prompt, "Make this thing look like Superman". In both cases it is the *user* that is initiating the process to create Superman, which DC owns. Should Photoshop be banned?
The problem is that the AI companies (Midjourney, Adobe, et. al) have knowingly and deliberately downloaded Superman into their systems -- so that when the user says, "Please do all the work of infringing on the Superman IP", the program does it.
It would be different if it was the user who uploaded Superman into the program and said, "Make the image of my father that we're editing look like this thing I uploaded".
I do not expect the tool to somehow police my content. Indeed, I would be upset if it did so. (
Courts will shut AI down (Score:2)
Not surprisingly, AI is an enormous threat to the value of intellectual property. Eventually the courts are going to shut down the pop-culture versions that compete with existing corporate media. Warner Bros. will win.
Re: (Score:2)
Horseshit. Artists, authors and musicians have lost cases over and over again on this.
[1]https://www.npr.org/2025/06/25... [npr.org]
[1] https://www.npr.org/2025/06/25/nx-s1-5445242/federal-rules-in-ai-companys-favor-in-landmark-copyright-infringement-lawsuit-authors-bartz-graeber-wallace-johnson-anthropic
Seems clear (Score:2)
From the summary, sounds like midjourney is a clear parasite that needs to go down....and definitely needs to go create its own content.
Re: (Score:3)
Seems clear that they're innocent, as you're allowed to train on other people's content for free.
Literally, not protected by copyright. At most you can claim trademark infringement for creations infringing trademark.
Incidentally, if you do accept that it's "parasitism" to learn from other people's content, can you name three artists who aren't parasitic that are alive today?
Re: (Score:2)
And you outlaw libraries, lest the content inside influences you in any way, and forms opinions that will be similar to what you read or saw. So.. museums, too.
Re: (Score:1)
It's not the training that is the issue. It's the production of essentially duplicate imagery that is copyrighted that is the problem.
Re: (Score:2)
Were these images requested by agents of Warner Brothers Discovery? Seems like that would damage their case.
Re: (Score:2)
Nice hallucination you have there. You AI fanbois are really the dumbest of the dumb.
Re: (Score:2)
Next they'll argue that submarines should be allowed to swim in the Olympics.
Re: (Score:2)
Probably.
Re: (Score:2)
> as you're allowed to train on other people's content for free.
> Literally, not protected by copyright.
There's no clear case law on this. A derivative work does not have to be a direct clear total copy. It can also copy some elements of the work. The LLM model weights may well be judged to be a derivative work of at least some of their training data. Of course the "AI" companies claim that's not true, but they do that because it's in their interest to claim that. We have repeatedly seen LLMs generate long passages of copyright works. Putting a filter in front of the LLM to hide that is in fact just making th
Re: (Score:2)
What's perfectly clear is how corrupt current copyright laws are. This just locking up our own culture behind corporate and classist paywalls. Most of this stuff should be in the public's domain already. What else is clear is how much this has cost us all and how much we lose everyday we don't have access to our own creativity and our works. We could be standing on the shoulders of giants instead of living in the barren empty wasteland of corporatocracy.
Re: (Score:2)
> What's perfectly clear is how corrupt current copyright laws are.
The Mickey Mouse copyright extensions made that pretty clear long ago. There's probably a case for saying that training LLMs should be allowed as long as the model weights and enough knowledge to use them are made public . The problem is if big corporations are taking all that knowledge and hiding it in house then that's worse than the current problems with copyrights.
Re: (Score:2)
fair use should allow for the use of our own culture to further our progress, and not be perverted in order to unfairly profit at the expense of progress