By: Bryan Sicard
As artificial intelligence (AI) becomes increasingly accessible to the general public, it is being used not only as an artistic novelty but has also shown itself capable of being used to fill voids in artistic markets. This has resulted in AI being directed to create poems and literature, songs, scripts, and visual arts: not just for personal enjoyment but for marketable, commercial purposes as well. As a result, trying to quantify AI’s artistic place in intellectual property (IP) law can seem like a daunting philosophical issue for attorneys. However, at least within the realm of copyright law, a recent case in the U.S. District Court for the District of Columbia can help illustrate that while concrete limits in copyright law exist, there are still many doors open for legal alternatives to protect works created in the AI-universe.
Stephen Thaler: a prolific litigator who previously failed at least twice to obtain patents for inventions created under similar circumstances, sought a copyright for an artistic image created by his AI software named the “Creativity Machine” (Machine) by virtue of his owning and programming the Machine. This piece, A Recent Entrance to Paradise, pictured above, a Kinkadian-like work of railroad tracks leading through walls of violets growing over bridges and buildings, was the subject of controversy with the U.S. Copyright Office (USCO). After the USCO denied Thaler’s application to copyright Paradise, he took the issue to federal court where Judge Beryl Howell appropriately ruled that the USCO was correct in denying Thaler his copyright. This is because the USCO followed the bedrock principle of copyright law that an artistic work’s author/creator must be a human being, and the work is driven by that person’s artistic direction. Thus, non-humans like the Creativity Machine cannot hold copyrights, nor can their human owners ever copyright these works.
Paradise clearly lacked human authorship when it was fixed as a tangible medium of expression because it was not truly by, or made under the direction, of Thaler. Rather, the true artistic actor in developing Paradise is the Machine, which might seem to contrast with the legal notion that artists (such as photographers) rely on technology to help them express themselves creatively. But the photographer simply uses the camera as a technological tool required for the mechanics of creating a photo. Without it, the photograph cannot be made, but the camera does not think creatively: it formulaically sorts through the positive and negative light to produce the image the photographer intended to create when pushing the shutter button.
Despite the ruling receiving a decent amount of mainstream press, Judge Howell reminded the parties and legal scholars that Thaler “is not nearly so complex in fact” as trying to “assess the originality of AI-generated works where the systems may have been trained on unknown pre-existing works.” Rather, this case is simply an issue of black letter law. Thaler may have contributed a minimal degree of creativity by writing the Machine’s software directions on what art to produce but did not independently create a product. He was instead entirely dependent on the Machine making artistic choices, such as tone, depth, and mood when creating the final image. If the Machine and Thaler were building back-and-forth off one another’s works like in a traditional artistic partnership: this would be a much more complex case with Thaler now having a better legal argument.
Importantly though, Thaler still has legal protections with Paradise: because he owns it by virtue of owning the Creativity Machine that created it. As a result, he still can most likely utilize trademark rights (using Paradise as a recognizable logo in interstate commerce) or can explore selling the original to collectors of non-fungible artwork. It’s crucial to remember that legal markets for AI-created artwork can still exist but those markets will simply have a more limited set of tools at artists’ and lawyers’ disposal than the commercial markets will allow. Attorneys can still draft and litigate contracts for issues surrounding sales, or trademarking AI-created artwork, in the same way a client’s rightful ownership of AI-software or directive coding gives them legal rights over the sale or use of their coding and the fruits it yields.
While Thaler v. Perlmutter at face value may seem to stifle AI’s role in future art-related IP matters: there are still many exciting options beyond just copyrighting an AI-piece. Appropriately, Howell’s ruling showed that while blackletter law means any piece’s potential copyrightability is not an unlimited right, the AI-generated artwork is not inherently unprotected. Rather, the limited scope of Thaler leaves so much future lawyering and litigable issues concerning AI-created art open in our ever-increasingly digital streams of commerce.
Student Bio: Bryan Sicard is a third-year law student at Suffolk University Law School. He is a staff writer for the Journal of High Technology Law. He received a Bachelor of Arts degree in History with a minor in Philosophy from Framingham State University in 2017.
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.
