Recent advancements in ‘artwork’ created by artificial intelligence have generated much excitement and discussion. Social media feeds have been filled with AI images generated from simple text based prompts. Users can produce slick, stylized, ‘artwork’ based on their favorite music lyrics, sci-fi fantasy descriptions, or random illustration requests. The visual or technical quality of artificial pictures has advanced to the point that they have now achieved commercial value. As the technology and image quality continues to improve, individuals will probably buy AI generated ‘art’ to suit their decorative or creative tastes. While corporations will most likely use the technology for advertisements, and publications might seek to use it for editorial illustrations. Yet the problem with this artificial ‘art’ revolution is the same as those that flow from most of our digital, technological advancements. The new technology just arrives on the scene of humanity, without any time for consideration about its impact on society, or discussion about its proper use.
It should be noted that I’m using the word ‘artwork’ in quotations when discussing the products of AI image generation. For I don’t believe that the images generated by the current iteration of AI programs should be considered art. True artwork is a product of self awareness and consciousness. A conscious creature perceives the world, applies the subjective nature of their mind, and creates an image or object derived from their mental processes. As of today, even the latest AI has not achieved self awareness or consciousness yet. AI image generators use processes that are more akin to those of a mechanical bread maker than a sentient creative being. For example, the AI image system ‘DALL-E 2’ uses a process called ‘diffusion’, which is a kind of digital programming that learns how to build images via dot and pattern recognition. Just as with the mechanical bread maker, there is no creativity at play. A human mind decides what ingredients to input into the machine, and out comes whatever bread (or images) they desire.
To be sure, the machine learning and algorithms that lie behind the latest AI image creation programs are more advanced than the straight forward engineering of a mechanical bread maker. Yet the basic principle remains the same. Both the mechanical bread maker and the AI image generator are tools designed to make a human task more efficient and easier. Both tools allow a task to be completed by a ‘non-expert’. Meaning the average person could perform the task of baking bread or creating an illustration without the help of a baker or artist. Yet this kind of technological efficiency comes at a price. Not a price that is measured by markets or interest rates, but one that is measured by human consciousness; or some would say, the human soul.
One day in the future, the improvement in self driving technology will remove the need for taxi drivers, limo drivers, or truck drivers. For the millions of people employed in these professions, this technological advancement will require them to find a new way to make a living. One would think that such an economic upheaval would only take place after meticulous planning and consideration. Yet ‘planning and consideration’ for such an event would only flow from a value (or moral) structure that has humanistic considerations. But in our ideologically atomized, techno-centric world, the dominant value structure is capitalism itself. And of course, capitalism is great for generating wealth and spurring technological advancement. But it was never intended to function as a wholesale, societal organizing system. Think of it this way. In Western countries where capitalism has become the dominant value structure, city development has moved steadily away from humanistic considerations. Instead of urban development that creates livable, multi-class neighborhoods; land and buildings are used as tools to generate the maximum amount of profit.
This is the terrain that AI automation will operate in; unfettered capitalism versus concerns surrounding human or social capital. There has been no discussion or planning about the human consequences of unleashing advanced AI image generators on the world. Not to mention the fact that every image generator had to be ‘trained’ on existing images. As of December 2022, two of the more advanced AI image tools (DALL-E 2 and Midjourney) have not made their initial program training datasets public. It defies belief that within datasets consisting of potentially billions of images, that there weren’t many that were used without permission. But what can be done now? The cat is out of the bag. Artists and photographers throughout history have unwittingly contributed to creating a technology that will try to render much of their future tradecraft obsolete.
Visual artists engaged in painting or ‘contemporary art making’ seem to be safe for now. That is until a fully conscious AI is able to apply paint to a canvas or create sculptures through subjective introspection. Yet for commercial artists and illustrators, there is a tsunami coming. Straddling the world of art and capitalism, they are especially vulnerable to an unsympathetic technological revolution. As AI image technology progresses, corporations and publications will find it hard to avoid the seduction of efficiency. Why deal with artists or handmade art when you can generate a slick illustration from your home computer or office? Just buy a one year subscription to an AI image generator like it was an Adobe graphics program, and generate any kind of image you can imagine.
Existing artists and illustrators can still fight to copyright their styles and designs. And I imagine publications like The New Yorker, or The New York Times will always value human made illustrations. But what will the world be like for the commercial artists of the future? How will they be able to compete with super advanced AI image generators that can conjure up any design or style they could ever think of? How will future generations of art directors or marketing directors not grow up to embrace the technology simply by circumstance? That is to say, every generation embraces the technology of its day like it is just normal, or ‘good’. A young person born today can never turn away from the smartphone or social media. Just as my generation could not have turned away from video games, cable television, or compact discs.
One phrase or line of thought I refer to often is the saying that the ‘medium is the message’. Coined by the great Marshall McLuhan, the phrase indicates McLuhan’s formulation that the contents of a given medium matter less than the overall effect a medium has on society. This sentiment seems more than accurate when thinking about AI image generation. The products, or creations of non conscious AI image generators will be nothing more than shiny, empty items. The images will be no more than trinkets of efficiency that ultimately grease the wheel of capitalism and business. Yet the overall impact of AI image generation mediums on society will be immense. And herein lies the problem. We are never able to plan or consider the effects of the technological mediums we unleash; because we become so hypnotized by their contents.
As the creator of the AI image generator, DALL-E 2, the company ‘OpenAI’ seems poised for Wall Street fueled, investor backed, financial success. Yet the primary driver of OpenAI’s growth is not DALL-E 2, but its artificial intelligence chat-bot or ‘smart’ search engine, ‘ChatGPT’. One could think of ChatGPT as an intelligent, conversational, Google search engine. Input any text prompt, or inquiry you can think of, and you will get an answer. It can even write ‘poetry’, or essays, which will no doubt send tremors through the writing community in the years ahead. In light of the disruption ChatGPT’s sister program DALL-E 2 will cause to human artmaking, I thought I might as well ask ChatGPT a few questions about the consequences that AI image generation could have on human society.
Here are the three questions I asked ChatGPT, and its responses:
What are some of the negative impacts on humanity as a result of art that is created by artificial intelligence?
There are a few potential negative impacts of art created by artificial intelligence that are worth considering:
Loss of jobs: As AI becomes more advanced, it may be able to create art that is indistinguishable from that created by humans. This could lead to the displacement of human artists, who may struggle to find work in a market that values human-created art less.
Ethical concerns: Some people may have ethical concerns about the use of AI to create art, particularly if the AI is generating content that is inappropriate or offensive.
Loss of human touch: Some people may argue that art created by AI lacks the human touch or emotion that is present in art created by humans. This could lead to a decrease in the value that society places on art.
Intellectual property issues: There may also be issues around who owns the intellectual property rights to art created by AI. If the AI is generating original works of art, it is unclear who should be credited as the creator and who should be entitled to any profits generated by the sale of the art.
What will happen to the profession of commercial illustration as illustrations created by artificial intelligence become more common?
It is possible that the use of AI to create illustrations could lead to the displacement of some human commercial illustrators, particularly if the AI is able to generate high-quality illustrations at a lower cost and faster turnaround time than humans. However, it is also possible that the use of AI could open up new opportunities for human illustrators. For example, human illustrators may be able to specialize in creating more complex or nuanced illustrations that are beyond the capabilities of current AI systems, or they may be able to use AI as a tool to help them create illustrations more efficiently. Ultimately, the impact of AI on the field of commercial illustration will depend on how the technology is developed and adopted.
Since human society revolves around capitalism, does it care about the negative effects that come with automation via artificial intelligence?
Capitalism is an economic system in which private individuals and businesses own and control the means of production and distribution of goods and services, and prices are determined by supply and demand. The primary goal of capitalism is to generate profit, and as a result, businesses may be motivated to adopt new technologies, including AI, if they believe it will help them to increase their profits. This can lead to automation, including the displacement of human workers by machines or AI systems, as businesses seek to reduce labor costs and increase efficiency.
While the adoption of AI and automation can bring many benefits, such as increased efficiency and productivity, it can also have negative impacts, including the displacement of human workers. It is up to society to decide how to balance the benefits and costs of AI and automation and to put in place policies and regulations that can mitigate some of the negative impacts.
This noble, final concept articulated by ChatGPT is one that society has already failed to heed when it comes to illustrators and commercial artists. If there was a slogan for AI fueled capitalism, it might be ‘shit out of luck’ followed by a middle finger emoji. Because that seems to be where all this is headed. Our future AI tools will emerge without little warning or consideration for their human consequences. Similar to the way social media emerged and crashed our national discourse. Except the consequences that flow from our AI tools will affect the very soul of the human beings they impact. They will take aim at what makes life meaningful for many, and what makes this human life, ‘human’ for some.