2

The AI-generated image is a misused tool that will change the artistic sector and its

commerce.

Maria Luna López Molina

Bachelor's Degree in Modern Languages with Emphasis in English

Faculty of Educational Sciences, La Gran Colombia University

Cesar Garcia-Herreros

March 1st, 2023

The AI-generated image is a misused tool that will change the artistic sector and its

commerce.

Artificial Intelligence (AI) has revolutionized the world of image generation and can

generate realistic images that are difficult to distinguish from those created by humans. This

technology has opened up a new era in the creation of images, as it provides an

unprecedented level of flexibility and creativity to artists, designers, and other professionals

who use images in their work, however, this mechanism is becoming obscure in terms of its

use since, at the moment, it has caused concerns among artists due to its easy accessibility

and mismanagement among users. This text will highlight why this tool isn't developing in

the best way, why it is revolutionizing modern art and how this affects the economic and

commercial artistic sector.

AI is being the biggest technological breakthrough of recent times, as humans are

trying to program and mechanize knowledge into machines that later can make conclusions

and creations on their own. This phenomenon has been very popular since the hong-kong

Sophia (the humanoid robot) project went viral on the internet and people started to talk

about artificial intelligences and how they will have an impact in the future.

These intelligences are evolving and becoming more and more present in daily life in

different fields, however, their elaboration is expensive since they require many resources to

be efficient, according to Zhang and Lu (2021) "AI is a knowledge project that takes

knowledge as the object, acquires knowledge, analyzes and studies the expression methods of

knowledge, and employs these approaches to achieve the effect of simulating human
 4

intellectual activities" so, an artificial intelligence needs a huge amount of information

(known as Big Data) and many training processes for its results to be more accurate.

Although it seems that these advances are supposed to be an advantage for workers

and consumers, a debate arose about artificial intelligences that generate images. Digital

artists began to notice that AI was imitating drawing styles characteristic of certain artists

such as Greg Rutkowski, who has worked for fantasy games like Dungeons and Dragons,

which is a problem since these AIs are even autocompleting images with pieces of

illustrations that are copyrighted.

Due to this, artists started a movement against the use of these pages as MidJourney,

Dall-e 2 and Stable Diffusion called "NO TO AI GENERATED IMAGES", publishing to

their social media an image to protest with the acronym AI and the sign of prohibited. An

interesting result was that in the social network Art Station (where well-known illustrators

upload their art with excellent quality), the movement was so immense that the images

generated by intelligence were affected with these symbols, confirming that the programs

were using art from the profiles of these illustrators.

The problem in this case is that being able to imitate the effort and talent of a human

with a free public intelligence, the work will be detracted or even worse, replaced. For

example, the case of Jason Allen, who won first place in the Amateur division of the Digital

Art category at the Colorado State Fair using an image generator and an editing program,

receiving a lot of criticism for not painting the picture by himself. Due to the recent

development of this tool, legally, the copyright does not apply to artificial intelligences,

which means that artists cannot take action on it.

As Morales (2020) explains, the essential requirement for a work to be considered art

is that it must be created by a human being and be original. This requirement can be

understood as subjective originality, which means, that the author hasn't copied someone

else's work; the second requirement is that it must be the product of human ingenuity and

creativity; and the third, it must be susceptible of being disclosed or reproduced, so taking

this into account, AI does not produce works in the legal sense. Using this "legal loophole",

many people have taken advantage of it in a debatable ethical-moral way, because as we saw

above, it is possible to commercialize it.

Understanding all these problems, it's normal that artists and designers feel worried,

because living from art has been quite complicated for years and it seems that this technology

won't favor them, but will take away their work. However, the terms 'Art' and 'Artist' are very

difficult to define, since throughout history this has been modified, having definitions such as

Thomas Aquinas "Art is the right ordering of reason" and Banksy "Art should comfort the

disturbed and disturb the comfortable" which arise at different times.

On one hand, the value of an artist is subjective, but on the other hand, in today's

working world, what makes an artist valuable in the market is the way he expresses his art

and the use of his techniques having a creative process. If artificial intelligence is imitating

these creative processes, it is replacing skills that only human beings could create. This is

how the competition begins, where the consumer and creator have a very important role to

play in preserving art as we know it.

This is not the first time that art is being revolutionized, since previously portraits

were made in oil to represent the image of a person, until the camera came to compete with

realistic painters. Along with many artistic movements, new forms of expression were born,

such as cubism and abstract art. This separated the market just like IA could do in the future,

creating new forms of art and expression.

Consumers should know that these AIs have limits and are generalized, that's why

somehow this art becomes repetitive, making the same mistakes and using certain patterns

that give away its use. That’s why the service will be divided between the "quick and easy

creation created by a machine" and "the traditional form of art by humans".

Artists must know how to use these resources responsibly and as a tool to enhance

their work, making the tedious processes of promoting themselves or serving their clients

more efficient, since this technology will continue advancing and it‘s not convenient for them

to refuse to use it. For example, many architects and animators have used these technologies

for mechanized processes such as 3D figure modeling, since by bringing these tools together

they save time in delayed processes to invest it in those that manually require a lot of

precision and time.

In short, artists will have to innovate and be creative enough to be chosen over a free

and instantaneous service that offers almost the same quality as them. Unfortunately, this will

not be possible unless a legal limit will be placed on these technologies to have a balance

between AI as a tool and the artists’ human skills. Moreover, it is necessary that the creators

of these intelligences (such as OpenAI) are aware that they are stealing the work that for

years has cost humans to create, so an interesting proposal would be to give the respective
 7

credits of the information they took, or in another scenario, pay for being able to take these

illustrations.

AI-generated images represent a significant advancement in the field of image

creation. The benefits of this technology are clear, including speed, flexibility, and high-

quality results, however, it is currently disadvantageous to the work of artists and designers

by using their copyrighted works without their consent, when this could be the tool that helps

renew art and the market among artists in a good way. Since this intelligence is developing in

more and more areas of life, the best option is to use it in a positive way, but this will only be

possible with the proper legal and ethical agreements to protect the artists' copyrights. AI is

not able to create art by itself, but humans can use it to make art.

References

Morales, A (2020). Derechos de autor e inteligencia artificial. Agnitio.

https://agnitio.pe/articulo/derechos-de-autor-e-inteligencia-artificial/

Zhang, C. Lu, Y. (2021). Study on artificial intelligence: The state of the art and future

prospects. ScienceDirect. https://doi.org/10.1016/j.jii.2021.100224

Bibliography

Crosier, S. (2002). Pablo Picasso, Cubism—A Revolution of Spatial Presentation in Artistic

Expression. Uc Santa Barbara.

https://escholarship.org/content/qt3wt7p9m2/qt3wt7p9m2.pdf?t=o0wtbp

Knight, W. (July 13, 2022) When AI Makes Art, Humans Supply the Creative Spark. Wired
 8

us. https://www.wired.com/story/when-ai-makes-art/

Lyn, A. (August 31, 2022). Someone had to be first': Pueblo artist criticized after AI painting

wins at Colorado State Fair’. The Pueblo Chieftain.

https://www.chieftain.com/story/news/2022/08/31/ai-painting-wins-at-colorado-state-

fair-pueblo-artist-explains-jason-allen/65466872007/

Nolan, B. (October 17, 2022). Artists say AI image generators are copying their style to

make thousands of new images — and it's completely out of their control. Insider.

https://www.businessinsider.com/ai-image-generators-artists-copying-style-

thousands-images-2022-10?utm_source=copy-

link&utm_medium=referral&utm_content=topbar

Wu, M. (September 2, 2022). AI-generated image wins art contest, ‘actual’ artists irate.

PopPhoto. https://www.popphoto.com/news/jason-allen-ai-artwork/