By: Annabelle Hentz
In the United States, copyright authorship may only be granted to works that are independently created by a human author and are sufficiently original. In a world where ever-changing technology continues to alter the scope of intellectual property and what works are inherently protected, artificial-intelligence-generated(“AI-generated”) images must be regulated. They must be reviewed to avoid infringement of protected works. Furthermore, they must be governed to ensure established licensing agreements and copyrights are being followed.
The United States Constitution created copyright law to “promote the Progress of Science and useful Arts.” Under current economic theory, copyright law seeks to achieve a balance between maximizing production of works of expression and their accessibility to consumers. Today, U.S. copyright law protects “original works of authorship fixed in any tangible medium of expression.” Copyright law recognizes eight categories of protected subject matters: (1) literary works, such as novels; (2) musical works, such as sheet music; (3) dramatic works, such as plays; (4) pantomimes (story and gesture expressed through movement) and choreographic works (particular dance style); (5) pictorial, graphic and sculptural works (two-dimensional and three-dimensional works of art); (6) motion pictures and other audiovisual works; (7) sound recordings; and (8) architectural works. The owner of a copyright contains the exclusive right to reproduce, distribute, publicly display and perform the work, and create derivative works. Registration of copyright, while not necessary for a creator to gain copyright protection, gives the copyright holder options regarding compensation if and when a lawsuit is ever brought.
Like many types of artificial intelligence, AI-art generators utilize various algorithms and datasets to obtain, organize, and reproduce information. The algorithms use information from preexisting art works, styles, and imagery. Some AI generators allow for the user to type in a description, and the AI tool will create a piece that resembles the description. Other AI generators allow its user to directly feed their own images to create new AI pieces.
In February 2023, the U.S. Copyright Office ruled that illustrations created by AI are not protected by copyright, given that only images created by humans can obtain copyright protection. This decision stems from comic book illustrations created by AI, utilized by author Kristina Kashtanova in her comic book Zarya of the Dawn. Kashtanova originally applied for copyright protection in September 2022, but did not disclose that the illustrations were not human-made. The Copyright Office further held that Kashtanova’s use of textual prompts into the AI generator did not qualify for copyright protection either.
Another interesting case regarding AI in 2023 is Getty Images v. Stability AI, where Getty Images sued artificial intelligence company Stability AI Inc., accusing them of misusing more than twelve million Getty photos to train its Stable Diffusion AI image generation system. Getty Images’ statement regarding the suit explained that they believe in and support the idea of training artificial intelligence to broaden creative endeavors, but Stability AI did not obtain any such license from Getty Images to use their images.
To regulate the production of artwork by artificial intelligence, the U.S. Copyright Office must strictly enforce the licensing agreements and copyrights of artworks with readily established intellectual property rights. While Getty Images believes and supports the idea of training artificial intelligence to broaden creative endeavors, their filing suit shows that they expect the rules to be followed, even if the creator of the infringing creation is not “human” and use of their images might not have been purposeful or malicious. Given that companies and individuals are beginning to implement AI for commercial use, AI access to licensed or protected works must be strictly regulated so that infringement does not occur. Alternatively, artists using AI within their own works, such as Kashtanova, must not purposely file their work as human-made when they in fact are not. Such filings would allow for inaccurate copyright protection so that falsely protected works’ rights would be given to claimless artists. Consistency and transparency regarding created works and established protections are necessary to regulate the future of copyright regarding both human-made and AI-generated art.
Artists, photographers, and copyright holders alike should change the copyright of their artwork to adapt to the ever-changing technology regarding the creation of artistic works. While it might be challenging to prohibit AI from accessing artistic works, stricter protection of artists’ works may help to prove copyright infringement if the work has been used. Further, established licensing agreements and copyrights must be enforced to protect artists’ work to avoid infringement.
Student Bio: Annabelle Hentz is a second-year law student at Suffolk University Law School. She is a staff writer on the Journal of High Technology Law and is a member of the Executive Board of the Women’s Law Association on the Diversity, Equity, and Inclusion Committee. Annabelle received a Bachelor of Business Administration in Marketing from the University of Massachusetts’ Isenberg School of Management.
Disclaimer: The views expressed in this blog are the views of the author alone and do not represent the views of JHTL or Suffolk University Law School.
