Increasing use of Artificial Intelligence (AI) has led to a series of challenges. Among them is the recent Intellectual Property (IP) litigation directed at AI generator companies. This has emerged as a key issue that requires careful consideration of both legal professionals and AI specialists.

The controversy over AI-generated art and its linkage to IP rights has been brewing for some time already. The rise of GANs (Generative Adversarial Networks) has allowed AI systems to create digital art that resembles traditional art styles, leading some to question whether such systems are essentially copying the work of human artists. The current lawsuit against Stability AI Ltd., reflects these concerns and could have far-reaching implications for the broader AI industry.

Getty Images, a stock photography company, and a trio of artists have initiated separate lawsuits against AI art generators - Stability AI Ltd., Midjourney Inc., and DeviantArt Inc. – for copyright infringement. Getty has accused Stability AI of “brazen infringement of Getty Images’ intellectual property on a staggering scale.” It has claimed that more than 12 million images have been copied by Stability AI from its database “without permission ... or compensation ... as part of its efforts to build a competing business.” Interestingly, some of the outputs generated by Stability AI include distorted versions of the Getty watermark, leading to a potential trademark infringement claim.

In yet another lawsuit, artists - Sarah Andersen, Kelly McKernan, and Karla Ortiz - have claimed infringement of rights of “millions of artists.” “Until now, when a purchaser seeks a new image ‘in the style’ of a given artist, they must pay to commission or license an original image from that artist,” states their attorney, Matthew Butterick. “Now, those purchasers can use the artist’s works along with the artist’s name to generate new works in the artist’s style without compensating the artist at all.”

While the plaintiffs in these cases argue that the AI-generated content is not transformative and constitutes copyright infringement under the law, the defendants argue that the AI-generated content is not a direct copy of the original works and is therefore not in violation of copyright law. They also contend that the AI systems are designed to create novel content that is distinct from the original works.

Determining whether AI art tools infringe on copyright law can be a challenging task. The vast databases from which these AI tools learn may contain images that are protected under the fair use doctrine. As reported by The Verge, the evaluation of potential copyright law violations requires an assessment of both the "inputs" (the images extracted from these databases) and the "outputs" (the images produced by the AI art generators). This is a complex matter that requires a nuanced approach.

With AI-generated artwork becoming more widespread, these disputes are likely to continue arising, and their outcomes could have significant implications for the future of IP rights and AI-generated works. While Some argue that AI-generated art can democratize access to the arts and allow for new forms of artistic expression, others worry that it could devalue human creativity and take away control over the use of one's own work.