Generative artificial intelligence has significantly changed the way digital images are created, enabling individuals to produce visual content quickly and at low cost through text prompts, without the need to hire digital artists. However, this newfound convenience comes with legal risks, particularly around copyright infringement.
The recent case of Andersen v. Stability AI, decided by the US District Court for the Northern District of California, clarified that both creators of AI programs and their users can be held accountable for copyright infringement if the AI-generated image resembles an existing copyrighted work. This case illustrates that AI systems, which build their knowledge from massive datasets of public images (known as training sets), can inadvertently duplicate or closely mimic these original works.
End users may fall into the misunderstanding that an AI-generated image does not infringe on copyrights simply because it appears novel. However, as the court noted, AI systems might not create entirely new images but produce results heavily influenced by the training set data, resulting in images that are “very similar if not identical” to copyrighted works. For example, an image of Marilyn Monroe generated in the style of Andy Warhol could potentially be very close to Warhol’s copyrighted work, leading to infringement.
To mitigate these risks, users are advised to conduct comprehensive copyright searches before using AI-generated images. This includes searching the US Copyright Office records, though its database can be cumbersome and lacks organization. Additionally, alternative methods such as examining the training set, reverse image searches, and seeking legal counsel are recommended. Notably, these efforts can demonstrate a lack of intent to infringe, enabling users to potentially invoke the “innocent infringer” defense under the Copyright Act, which can reduce statutory damages in case of a lawsuit.
In summary, while AI-generated images can offer new efficiencies and creative possibilities, they also pose substantial risks of copyright infringement. Legal professionals and companies leveraging such technology must exercise due diligence by conducting thorough searches and considering potential legal implications to avoid unwelcome copyright litigation. For more detailed insights, see the full article on Bloomberg Law.