Artificial Artist: Whose Art is it Anyway?
In the rapidly growing field of Artificial Intelligence, the birth of AI-generated art sparks important questions about the nature of creativity, ownership rights, and ethics. As AI technologies progress, producing increasingly sophisticated and aesthetically pleasing works of art, the traditional understanding of creativity as a distinctly human attribute becomes increasingly murky. This paper addresses issues surrounding AI-generated art, aiming to discuss the complexities of creativity in the digital age, navigate the murky waters of ownership rights in AI-generated creations, and define the ethical considerations of AI as a creative entity. By examining these themes, this research seeks to shed light on the evolving landscape of art production and consumption in an era dominated by technological innovation.
The first major point is to establish AI's ability for creative expression. To do this we must first define what constitutes art and creativity. There are differing opinions on the defining characteristics of the creative process, but many scholars agree in defining art in a way that depends on the creator's mental input and expression. One expert emphasizes the creator's intention when defining art and creativity, although it is not a necessary condition for an object to be considered a work of art. "It states that a work of art is [ . . . ] either an arrangement of conditions intended to be capable of affording an experience with marked aesthetic character or (incidentally) an arrangement belonging to a class or type of arrangements that is typically intended to have this capacity" (Mikalonytė 3). They go on to say that another traditional defining characteristic of art is the creator's expression of their inner emotion, which would potentially exclude AI as artists because they don't have emotions to express. Art must be intentional and Mikalonytė states "the folk concept of intentionality consists of five elements: a desire for an outcome, a belief about the action leading to that outcome, an intention to perform the action, awareness of fulfilling the intention while performing the action, and the skill to perform the action" (Mikalonytė 4).
With those definitions of art, we must then determine if Artificial Intelligence is capable of diverse, creative works or if it can only produce work based on preprogrammed parameters. There was an experiment conducted to gauge people's considerations towards the creation of art. The participants were shown pieces of art and were presented with a scenario surrounding the creation of that corresponding piece. They were then asked a series of questions on whether they believe the creation to be art and whether they consider the entity responsible (both human and AI creators) as an artist. The findings determined that the participants were more likely to consider the works created by humans to be art and they were less inclined to consider non-human, or AI, to be artists (Mikalonytė 6). One expert, R.K. Sawyer, explains that while Artificial Intelligence programs are capable of "crunching mounds of data and identify patterns invisible to the human eye, they still cannot master everyday creative skills" (Cheng 3.1) while another expert counters that sentiment by providing evidence to the contrary. This expert, M.A. Boden, takes many qualifying characteristics of the creation process, which are primarily human in nature, and gives different examples of how AI can meet those qualifications just as well as humans can (Cheng 3.2).
Now that we have addressed if AI is capable of creativity and producing art, we must now consider the legal complexities surrounding the ownership and rights of the artwork generated by Artificial Intelligence. The current legal processes in place for ownership rights of AI-generated works are still in their infancy. If an AI program creates artwork, does the computer program own the copyrights or does the person who created the program's code and algorithms have the right to claim ownership of the work. One current consideration in AI-generated artwork ownership is the amount of human involvement in the creation of the artwork. If someone tells the AI program exactly what to create and sets various parameters, they play a direct role in the creation of that artwork. According to current US copyright laws, the human who "commissioned" the artwork would own the rights but must cite the AI program used in the creative process (Hristov 5). More autonomous AI programs, which have little to no human guidance outside of the initial creation of the program itself, may have their source code copyrighted, but the works created by the AI program cannot be copyrighted; they fall into the Public Domain (Hristov 6).
The use of AI poses several challenges in determining the authorship and attribution in AI-generated works. In a survey conducted with various AI, copyright, and other related experts, some expressed their beliefs that the works produced by AI should belong to the creators of the AI program and that they initiated the creation of that work in the same way an employer might commission work to be done by an employee (Hristov II v.). Hristov adds his thoughts on the matter of treating writings produced by AI as works for hire, which would allow the writing to fall to the human creator of the program instead of being in the public domain. He also says that AI is incapable or ill-suited to fulfilling the obligations associated with copyright protection and copyright infringement (Hristov II vi.)
There are problematic implications of AI-generated art in relation to intellectual property rights. One such implication is that non-human creators cannot be held legally responsible in a court of law, which creates many questions on the eligibility of AI having intellectual property rights. Allowing a non-human entity to have intellectual property rights opens the legal system up to many cases of abuse of those laws for the lack of a legally responsible party. The case Naruto v. Slater questioned the rights of photos taken by monkeys, and the courts ruled that even though the monkey was directly responsible for the creative work, they could not be considered the author for the purposes of the law (Hristov 19).
This paper has addressed and provided information about the creative expression of Artificial Intelligence and whether the works generated by AI are considered art, and whether AI programs are comparable to artists. It has also expressed current legal challenges related to ownership and rights to AI-generated art and the current standing of laws in place defining intellectual property of AI works. The ethical implications surrounding AI's integration into the artistic landscape and what it means moving forward for human artists and authors are still in its infancy. My hope is that there be future research and further policy development for AI-generated works.
Works Cited
Cheng, Mingyong. "The Creativity of Artificial Intelligence in Art." MDPI, Multidisciplinary Digital Publishing Institute, 6 Apr. 2022, mdpi.com/2504-3900/81/1/110.
Hermerén, Göran. "Art and Artificial Intelligence." Cambridge Core, Cambridge University Press, 4 Apr. 2024, cambridge.org/core/elements/art-and-artificial-intelligence/CE565B67A2BD0416A1CB359AA9E659FF.
Hristov, Kalin. "Artificial Intelligence and the Copyright Dilemma." IP Mall, 2017, ipmall.info/sites/default/files/hosted_resources/IDEA/hristov_formatted.pdf.
Hristov, Kalin. "Artificial Intelligence and the Copyright Survey." ResearchGate, 1 Apr. 2020 researchgate.net/publication/342961336_Artificial_Intelligence_and_the_Copyright_Survey.
Li, Yueen, et al. "Research on Artificial Intelligence Ethics in the Field of Art ..." IOP Science, IOP Publishing Ltd, Sept. 2020, iopscience.iop.org/article/10.1088/1742-6596/1673/1/012052/pdf.
Mikalonytė, Elzė Sigutė, and Markus Kneer. "Can Artificial Intelligence Make Art?: Folk Intuitions as to Whether AI-Driven Robots Can Be Viewed ..." ACM Transactions on Computer-Human Interaction, 8 Sept. 2022, dl.acm.org/doi/pdf/10.1145/3530875.
