AI-Generated Art: The Intersection of Technology, Creativity, and Ethics
Art has always been a means of expressing ideas, emotions, and experiences that transcend language and culture. However, the ethical issues surrounding art creation have been a constant source of debate throughout history. In recent months, questions of ownership and originality in art have once again become topics of intense discussion, thanks to a new technological breakthrough that has revolutionized the art world: AI art generation. The use of artificial intelligence to create art has raised questions about the role of the artist, the authenticity of the work, and who owns the rights to it. As AI-generated art gains popularity, it is essential to consider the ethical implications of this new medium and its impact on the art world.
AI art is a form of generative art that is created using algorithms and machine learning techniques. To create AI-generated art, artists feed a computer program with vast amounts of data, such as images, sounds, and texts. The program then uses this data to generate new and original artworks, which can take the form of images, videos, music, or even entire virtual worlds. The resulting art is often unpredictable. The computer program is capable of making decisions that are beyond the control of the artist. However, this also raises concerns about the impact of AI art on the traditional roles of artists and the art world as a whole. The lack of human intervention in the creative process challenges the concept of authorship and originality, while the use of algorithms to generate art raises questions about the authenticity and uniqueness of the resulting works. These ethical dilemmas have led some to criticize the use of AI art as a gimmick, rather than a legitimate form of artistic expression.
As secular humanists, it is our duty to approach new and disruptive technologies with both an open mind and a critical eye. When a technology like AI-generated art raises real and significant ethical challenges, it is essential that we take a deep examination of that technology and not resort to dogmatic assumptions based on our past experiences or flawed understandings. We must consider the impact of AI art on our society, culture, and values, and ask ourselves difficult questions about the role of art in our lives and the implications of using machines to create it. Only through thoughtful and nuanced discussion can we hope to navigate the ethical dilemmas posed by AI art in a way that upholds our humanist values and promotes the common good.
One of the most common objections to AI-generated art is the claim that it constitutes theft when the training data set uses artwork without the artists’ permission. Some argue that feeding an AI algorithm with copyrighted images or music, for example, is a violation of intellectual property rights. This objection is based on the assumption that AI-generated art is a direct copy or reproduction of existing artworks, rather than a new and original creation. However, this assumption is not entirely accurate. It is virtually impossible for any AI-art algorithm to generate an exact copy of an existing piece of art, even if that piece was used in training the model. While AI-generated art may be reminiscent of existing styles or works, it is still an original creation. This is similar to how human artists can also replicate the style or subjects of other artists, and as long as they are not doing so deceptively, it is typically not against the law. In fact, it is regularly encouraged by many artists. A quote often attributed to Picasso applies here - "Good artists borrow, great artists steal." Nonetheless, it is important to acknowledge the potential ethical implications of using copyrighted material in AI-generated art, and artists should be mindful of copyright laws and seek legal advice when necessary.
The most ethical approach that we might strive for would be to appropriately label AI-generated art whenever possible and to use AI services where permission has been granted by all artists whose work was included in the training model. Unfortunately, it will be impossible to prevent deceptive use of this technology, just as it has been impossible to prevent human artists from creating skilled forgeries. Laws concerning deceitful use of AI are sure to evolve over the coming years to deal with this real and worrisome threat, but it is clear that not every use of this technology will constitute an infringement of someone’s copyright protections.
Another common objection to AI-generated art is that it devalues the work of original artists by generating similar works much faster and cheaper than humans can, thus lowering their economic value. This is a common theme that arises with every technological innovation that reduces the need for human labor. However, it is important to consider that the value of art is not always correlated with the amount of effort involved in producing it. The worth of art is often a subjective and fickle thing, with some pieces selling for millions despite being created with minimal effort, while others that required years of effort might never be monetarily valuable.
While there is no doubt that AI-generated art can produce imitations of existing art styles at a fraction of the cost and time it would take for humans, this doesn't necessarily mean that it will devalue the original works or reduce the income of all artists. Master carpenters have been able to find a niche despite the availability of mass-produced furniture. Live subject painters have managed to innovate and prove the value of their work despite the invention of photography. There will likely always be a market for human-authored works of art.
It is important to recognize that the monetary value of art is often tied to the story behind the piece, perhaps even more so than the perceived quality of the piece itself. An original work that tells a compelling story or has a significant historical or cultural value will always be highly valued, regardless of whether it was created by a human or an AI algorithm. It is possible that in the future, the story behind AI-generated art may even become part of what makes it valuable.
Another common criticism leveled at AI-generated art is that it is simply "bad". This objection typically stems from the limitations of current AI technology, which struggles with creating certain elements of art, such as realistic human hands. However, it is important to note that AI technology is advancing rapidly, and it is likely that these limitations will be overcome in the near future. In fact, AI-generated art is already improving at an exponential rate, and some pieces are becoming increasingly difficult to distinguish from those created by human artists (or even from real photographs). As the technology continues to develop, it is highly probable that AI-generated art will eventually become indistinguishable from art created by humans. That may seem like a frightening eventuality to some, but that has tended to be the response throughout history of those who have difficulty imagining further innovation beyond the status quo.
Perhaps the most significant worry with regards to the advent of AI-generated art is the potential for it to be used to deceive or manipulate people. We’ve already discussed how this technology could be used to create art forgeries, but AI also has the ability to create hyper-realistic images and videos that can be used to create convincing but fake news, or even be used in scams or other malicious activities. While this is a serious concern, it is not an entirely new one. The invention of technologies like Photoshop decades ago made it possible for convincing fake images to be created. The difference is that the speed and accuracy of the image generation tools may simply overwhelm our capacity to validate the truthfulness of the created works. Unfortunately, the technology is already far too accessible to truly restrict its growth and adoption by malicious actors. We will need to be discerning of how we apply laws and restrictions aimed at lessening the negative impacts to our social order.
While these are all valid concerns regarding AI-generated art, it's important to also consider the potential benefits this technology could bring to the art world and to the larger human experience. From creating new forms of expression to increasing accessibility to art, AI-generated art has the potential to revolutionize the way we create and appreciate art. Let's explore some of the positive aspects of AI-generated art and how it could change the world as we know it.
To start, the term "AI-generated art" is somewhat misleading, as it implies that the technology is capable of replicating the abstract thought and creativity of a human brain. In reality, the algorithms behind AI art are simply complex mathematical models that are capable of pattern matching and generating data. Similar models have been used in the film and television industry for decades to simulate realistic lighting and physics, and to create stunning visual effects that were once impossible to achieve. As a result, the use of these mathematical tools has enabled a new era of creativity and innovation, allowing artists to bring their ideas to life in ways that were previously unimaginable. Studios like Pixar and Industrial Light and Magic are revered for their artistic achievements where it is undeniable that the work would not have been possible without the aid of these computer models. What might have taken weeks or months of painstaking human labor can now be accomplished in a matter of minutes and to higher degrees of fidelity.
Ironically, similar objections to the use of these technologies were levied decades ago when it was feared that computer generated imagery would destroy the artistic integrity of filmmaking. The groundbreaking feature film TRON (1982) was famously disqualified from any oscar nominations in special effects, because the Academy felt that using computers was a form of “cheating”. Today, tools with orders of magnitude more complexity are used almost invisibly to assist in the creation of some of the most respected and valued works of art. Over time, these tools have become normalized after we see what they can achieve and how they can augment the human creative process.
AI-generated art has the potential not just to augment creative workflows, but it may also make art more accessible and inclusive for people with physical or mental impairments. Traditional forms of art can often require a high degree of physical dexterity or fine motor skills, which can be a challenge for individuals with disabilities. However, AI-generated art tools can provide an alternative means of artistic expression that does not require the same level of physical ability. For example, AI tools that can translate speech or text into visual art or music can allow individuals with physical impairments to create art or music in a way that was not previously possible. Additionally, AI-generated art could be used as a therapeutic tool for individuals with mental health conditions, providing an outlet for expression and creativity that may be difficult to achieve through traditional methods. The fact that many of these AI tools are already free or extremely inexpensive means that they are also a way to make art more available to those with lower incomes. You could generate a piano solo without a piano. Or generate beautiful oil paintings without the expensive brushes, canvases, and paints. By breaking down physical, cognitive, and economic barriers to artistic expression, AI-generated art has the potential to democratize art and make it more inclusive for everyone.
As with art itself, it seems that our perception of AI-generated art will be largely subjective and unique to each individual. It is true that dangers and challenges exist. The question is, will these concerns be met with equal or greater positive impacts? Will artistic skill be devalued while our society struggles to differentiate truth from fiction? Or will artistic variety and appreciation be greatly increased due to a dissolution of physical and mental barriers? As humanists we need to be constantly vigilant of how any new technology impacts the human experience and our connected world. But a steadfast rejection of innovation out of fear has rarely benefitted the progress of humankind.
Author’s Note: We have only skimmed the surface here of what possibilities exist with AI technologies. For example, there has been no mention of the capabilities of AI to generate lengthy text from short prompts and outlines. Entire articles can be created with relatively low effort. In my exploration of this topic I decided to experiment with this technology. Roughly 80% of the above text was written by an AI called ChatGPT. Even I was surprised by just how much of the text I was able to generate while still maintaining a feeling of ownership over the content, tone, and emotion. It felt like a highly competent writer was helping me express my own ideas in compelling new ways that I was able to further edit and modify to my liking. I worked in collaboration with a computer and it was exciting, mystifying, and sometimes a little bit scary.