Artificial intelligence has rapidly advanced over the past few years, with generative AI emerging as a particularly powerful tool that can generate text, images, and art autonomously. While this technology has numerous applications in fields ranging from medicine to design, it has also raised concerns about the potential infringement of human creators’ copyrights.
With generative AI platforms such as Stability AI allegedly using copyrighted material to generate new works without the creators’ consent, there is growing concern among artists, photographers, and other content creators about the future of their industry.
In this article, we will explore the legal disputes that have arisen between human creators and generative AI companies and the possible paths forward to resolve the tension between human creators and technology’s unrelenting march forward.
The Generative AI Litigation Landscape
In the last year, several high-profile lawsuits have been brought alleging generative AI that is trained on copyrighted material is violating creators’ copyrights when it generates, for example, new images.
In one lawsuit, a trio of artists—Sarah Anderson, Kelly McKernan, and Kara Ortiz—filed suit in California against Stability AI and Midjourney, two organizations that generate images. The artists allege that these AI generative platforms have scrubbed the internet for five billion images and used it for their tools without acquiring consent from the creators. The artists contend that these images infringe on their copyright.
The second case involves Getty Images and a Delaware lawsuit against Stability AI. The allegations are similar to the California lawsuit. Getty also alleges Stability AI is infringing on the copyright of artists by web scraping 12 million images to train its AI generative platform.
These cases, along with others that are likely to emerge, will have a significant impact on the future of generative AI and its relationship with human creators. As the legal landscape continues to develop, it is important to consider the potential ramifications for artists, photographers, and other content creators, as well as the broader implications for the use and development of generative AI.
In these and other cases, one of the key legal issues is the concept of “fair use.”
Defining Fair Use
Fair use is a key consideration in generative AI legal disputes, as it allows individuals and organizations to use copyrighted material without obtaining permission from the creator. Whether fair use is a feasible defense for generative AI platforms like Stability AI will depend on whether the works generated are transformative.
Transformative use involves using copyrighted material in a way that adds something new and significant, rather than merely copying or reproducing the original work. In some cases, courts have found that using copyrighted material to create new works can be transformative and thus fall under fair use. For example, in the 2021 Supreme Court case Google LLC v. Oracle America, Inc., the Court found that using Java SE code to create a new operating system was transformative and fell under fair use.
Stability AI argues that its use of copyrighted material is transformative, as the generated works are not exact copies of the originals but rather variations that add something new and significant. The Second Circuit case, Authors Guild, Inc. v. Google LLC, may lend support to this argument. In this case, the court found that creating a catalog for searching digital books and sharing snippets of them without permission was considered transformative use.
On the other hand, the artists who are suing Stability AI argue that the generated works are not transformative as they do not add anything creative to the original expression. In some cases, the original watermark of the creators even appears in AI-generated images, which supports the argument that Stability AI and similar organizations are merely copying. Ultimately, the question of whether the use of copyrighted material in generative AI is transformative and thus falls under fair use is fact-driven and will depend on the specific aspects of each case.
As the legal disputes continue, it will be interesting to see how courts apply fair use to generative AI and whether any new legal and/or business frameworks emerge to address the unique challenges posed by this technology.
Possible Paths Forward
As the debate over generative AI and copyright continues, there are several possible paths forward to resolve the tension between human creators and technology.
One potential solution is to create a new legal framework specifically for generative AI, which could help clarify the rules surrounding the use of copyrighted material in AI-generated works. Such a framework could potentially include new regulations, guidelines, or licensing requirements that specifically address the unique challenges posed by generative AI.
Another solution is to develop new technologies that can protect creators’ rights in the age of generative AI. For example, some experts have proposed using blockchain or other decentralized technologies to create a transparent and immutable record of ownership for creative works. This could help ensure that creators are properly compensated and credited for their works, even as generative AI continues to evolve.
However, based on past business precedent, it is likely that licensing deals negotiated between generative AI platforms and creators will be one of the primary paths forward. With a licensing agreement, creators can have rights in negotiating how they want their intellectual property to be used and how they will be compensated. Similar to the licensing deals negotiated between music services like Apple’s iTunes and Spotify with record labels following the downfall of Napster, these agreements can help ensure that artists are paid royalties for their work, while allowing generative AI platforms to continue to innovate and create new works.
Ultimately, finding common ground through licensing agreements or other solutions will be essential for ensuring that generative AI and human creators can coexist, while still protecting the value and integrity of existing intellectual property. As the technology continues to evolve, it will be important for all stakeholders to work together to find a way forward that balances innovation with the rights of creators.
Conclusion
The legal disputes arising between human creators and generative AI companies highlight the need for a new framework that can balance innovation with the rights of creators. While the definition of fair use will be central to these disputes, it is clear that new technologies and regulations will also be required to address the unique challenges posed by generative AI.
Possible paths forward include negotiating licensing agreements between creators and generative AI platforms, developing new technologies to protect creators’ rights, and creating a new legal framework specifically for generative AI. Whatever paths forward are chosen, it will be essential to find a way for generative AI and human creators to coexist—protecting the value and integrity of existing intellectual property while continuing to encourage technological innovation.
