In a world where the field of art is dominated by Artificial Intelligence (AI), Picasso’s genius would be mere code. As recent advancements in generative AI have made image generation widely available to the public, everyone has access to creating their own digital media. The consequences of this are that individuals are now capable of creating an art piece in a blink of an eye, without any prior knowledge or skill in art required. One could argue that these ‘advancements’ could empower individuals, who previously were not able to realise their creative potential. However, things are not as simple as they may seem on first glance, and not only may the meaning of art be devalued, the entire process of the creation of an artwork may be dehumanized. Therefore, this article will argue for the following statement:
AI-generated art is not real art since it is not rooted in conscious thought, and is therefore dehumanising and devaluing original art.
Ever since its advent, AI-generated digital media has been a controversial topic. Commonly discussed issues concern copyright, income stability for artists, misuse (like deepfakes or hate-speech) and sampling bias. However, one of the most prominent issues within the art world is whether or not AI art should be held in the same regard as human-created art. Can we truly see AI art generated by a prompt as the creation of art? Can we allow for human-created art to share the same definition, even when AI is not able to interpret the meaning of the piece it created?
The importance of distinguishing between AI and human created artworks
Arts and crafts are, as put by artist Leonard Baskin, distinctly human hobbies and fascinations. By automating the process of creating art with AI, we are dehumanising the process of creating a piece. The meaning of an art piece may be lost, as the creator of the piece (the AI) is not able to grasp the subtleties of a human thought. Often when a thought is expressed as a prompt, AI is not able to generate the exact image the person had in mind. Even though the picture can come very close to the original thought, AI is still unable to comprehend the extensive details and emotion behind an idea. For example, when asked to generate the image of a nerd, the AI model DALL-E was only able to generate images of a nerd with glasses. When asked to remove the glasses from the image, DALL-E was unable to fulfil the request. This may indicate that AI as it currently stands is unable to create the exact piece of an idea that someone has in mind.
Furthermore, the creation of art with AI can be done by anyone at anytime at a fraction of the cost of time and effort. As a result, original artists might be outcompeted as they are not capable of creating such large quantities of content. In the current age of social media, where the amount of online content created is at an all time high, the pressure for artists and small businesses increases to stay in the public eye and promote themselves in the vast oceans of content available. While the creation process of a human-created piece may take significantly longer than an AI-generated piece, the value relies on the original thought behind it. If we were to allow for AI to roam free, generating content at extreme rates, we would quickly see the surge in AI content as it learns what the most engaging art pieces are. Original artists will not be able to keep up with this rapid surge. Are we willing to miss out on original thoughts and creations, only to replace them with meaningless content that happens to be optimally engaging for a certain target audience? If we allow the dehumanisation of creating art to continue, we will reach a point where human artists are unable to break through the AI-barrier, and human-created artworks will not find the same audience and appreciation as before.
Additionally, the artwork on which AI is trained is often used without explicit consent from the artists. If an artist has a distinct style within their pieces, and this style is culturally or personally significant to the artist, the AI may copy elements of the piece and use it outside of the original context. As the significance of the artist’s original work is restructured and recombined with non-relevant elements by the AI, the original significance of the artwork is lost in the new creation. Therefore, AI may also contribute to the devaluing of original artworks.
The differences in creation processes of art
But wait, a critic might say, what exactly is the difference between a human and an AI-system creating art? After all, they both draw inspiration from existing art and create new art based on their experience. However, the problem with this line of thinking is that AI is not capable of creating fundamentally new meaning. The only thing it is capable of is abstracting old (human) art to create something (new) that resembles that. Additionally, research1 shows that people tend to prefer human artists over AI, even when they can not tell the difference. To answer the question if AI-generated art is real art, let us delve deeper into what specifically gives meaning to art.
What gives meaning to art?
Art is a highly subjective domain. Quantifying the meaning of a specific art piece is therefore not possible. We can, however, reflect on the process of how meaning is substantiated. In the context of AI-generated art, a possible view might be that it is not the (AI-)artist who gives meaning to the piece, but rather the way it is perceived (by a human). While this argument is valid, although still subjective, we believe it is flawed. Perceiving art can be thought of as an interaction. The art has some influence on the viewer (or not) and the viewer naturally tries to interpret this based on the circumstances and cultural background in which the piece was created. Given the fundamentally different way of AI creating art, it is not possible to interpret it since we can not compare ourselves to the creator. The opposite holds as well, since an AI-system is not physically part of society, it has no first hand experience, thus it can not form opinions or beliefs of their own. If the creator had no intent when creating a piece there is no meaning in interpreting it, therefore AI is incapable of creating meaningful art. This disconnect between the actual world and the AI-verse, is the root of the problem.
What now?
Should we forbid the use of AI in art, fund the original creators or trust that the lack of new meaning will drive viewers to original creators? Since banning AI is both impractical, unnecessary, and given the existing preference of human-created art, we plead for transparency, specifically regarding the use of AI, in the creation process of art. If we do not prioritise this issue, we risk to abnegate our human self in the process. While AI already has revolutionary capabilities, this is the critical moment for society to think about why (and why not) we want to implement this technology and to make sure that it does not defeat the purpose of creating art: connecting with other people.
- Bellaiche, L., Shahi, R., Turpin, M. H., Ragnhildstveit, A., Sprockett, S., Barr, N., … & Seli, P. (2023). Humans versus AI: whether and why we prefer human-created compared to AI-created artwork. Cognitive Research: Principles and Implications, 8(1), 42. ↩︎
