Getty Images’ recent lawsuit against image synthesizer Stability AI introduces so many talking points — and even more questions — into the IP law conversation:
What is fair use? When does the use of AI constitute copyright infringement? Could AI be regulated or restricted in the IP process through government intervention? Should copyright holders be on the lookout for copyright-violating AI algorithms, and what can they do about it?
If the complaint, filed in February, achieves anything, it’s to set a new precedent as technology advancements continue to change and shape the world of intellectual property law.
The Seattle-based stock photography and visual media company claims that the AI startup misappropriated more than 12 million Getty-copyrighted photos to train its Stability Diffusion AI image recognition software. The Stability Diffusion computer generates pictures by studying millions of human-created images that tell the software, through a prompt, what they should look like.
Getty is seeking $1.8 trillion in damages, or roughly $150,000 per photo.
What’s the basis of the lawsuit? What are the implications and outcomes? And will it change the IP legal landscape?
Getty Images’ Stable Diffusion Lawsuit: The Details
“This case arises from Stability AI’s brazen infringement of Getty Images’ intellectual property on a staggering scale,” Getty Images alleges in its complaint, filed in the U.S. District Court for the District of Delaware. “Upon information and belief, Stability AI has copied more than 12 million photographs from Getty Images’ collection, along with the associated captions and metadata, without permission from or compensation to Getty Images, as part of its efforts to build a competing business.
“As part of its unlawful scheme,” the complaint continues, “Stability AI has removed or altered Getty Images’ copyright management information, provided false copyright management information, and infringed Getty Images’ famous trademarks.”
According to the filing, rather than attempting to negotiate a licensing agreement with Getty for the use of its content, Stability allegedly and unlawfully copied scores of its images to populate its learning model, preventing Getty from fulfilling its business model of generating revenue from image licensure. The allegation includes several Stable Diffusion-generated images side by side with the copyright-protected Getty Images photos — some of which contain a distorted Getty Images watermark to indicate where the images were sourced.
“Stable Diffusion,” it states, “at times produces images that are highly similar to and derivative of the Getty Images proprietary content that Stability AI copied extensively in the course of training the model. Indeed, independent researchers have observed that Stable Diffusion sometimes memorizes and regenerates specific images that were used to train the model.”
Founded in 2020 by a former hedge fund executive, Stability AI is valued at $1 billion, having recently raised more than $100 million from venture capital investors.
When IP and AI Collide: Issues and Implications
The case highlights the growing importance of AI in content creation and distribution, and the challenges that arise at, in essence, the intersection of copyright laws and technology. As AI algorithms become more sophisticated, they are increasingly being used to create and distribute content, which raises important questions about who owns the rights to this content.
Fair Use
Stability AI may likely respond that it hasn’t violated any copyright or trademark laws because the images were used under the fair use doctrine, which allows limited use of copyrighted material without obtaining permission from the copyright owner. It could be argued that its use of the images was transformative, as it was only using the images to train its AI algorithms and not for any commercial purpose. Read more here about the four factors of fair use in copyright law.
The fair use doctrine has traditionally been used to allow for limited use of copyrighted material in certain circumstances. But with Getty Images and Stability AI, it’s unclear how it applies to the use of AI algorithms to create and distribute content. As AI technology becomes more prevalent, courts will need to grapple with these issues and determine how copyright laws should be applied in the context of AI-generated content.
Copyright Infringement
Another issue is the liability of AI systems themselves for copyright infringement. Under current U.S. law, copyright infringement can only be committed by a human actor. AI systems are becoming more sophisticated and autonomous, and it may become more difficult to determine who is responsible for any infringement that occurs. This could lead to situations where companies are held liable for the actions of their AI systems, even if they had no direct involvement in any infringing activity.
The Getty/Stability case also highlights the need for copyright owners to be vigilant in protecting their intellectual property in the age of AI. Algorithms may eventually be able to create content that is nearly indistinguishable from human-created content. Copyright owners will need to be vigilant in protecting their IP and enforcing their rights against infringers, whether human or machine.
An Evolving Relationship
Does either side have a case? The outcome of this lawsuit could have significant implications for the development of AI technologies and the protection of intellectual property rights in the digital age. If Getty Images is successful in its lawsuit, it could set a precedent for other companies to take legal action against AI companies that use their copyrighted works without permission.
On the other hand, if Stability AI prevails, it could create greater uncertainty around the ownership of intellectual property in AI systems, and make it more difficult for copyright holders to protect their works.
Regardless of the verdict, one potential outcome of the case is that it could lead to new regulations or laws governing the use of AI in content creation and distribution. As AI technology becomes more prevalent, there may be a need for clearer guidelines on how copyright laws apply to AI-generated content. This could include new regulations on how AI algorithms can be trained using copyrighted material or new laws governing the ownership of AI-generated content.
Protect Your Intellectual Property
Getty Images’ predicament happens every day on small and large scales to content creators across the country. That’s why it’s important for individuals and enterprises alike to copyright their works, and remain vigilant in protecting those rights with the proper legal help.
Patterson Thuente’s experience, know-how and prestige in IP law have helped creative and inventive professionals, in Minnesota and worldwide, protect and profit from their ideas without anybody else doing so unlawfully.
How can we help you with your IP legal needs? Contact us today and get your ideas owned.