Examining the intertwined history of art and technology
Recently, humanity has witnessed a rapid evolution in the quality of artificially-generated images. From humble six-fingered and disproportionate beginnings, AI images have become nearly indistinguishable from the hand-drawn. As AI grows more sophisticated, it continues to creep into our everyday lives. Businesses are adopting AI logos and clipart, and our social media feeds are flooded with generated images in a soulless Studio Ghibli-like art style. We’ve reached a point where intimate reflection about whether what we see is real has become necessary.
When discussing the future of AI with peers, the period of the invention and the international spread of camera technologies comes to mind—especially its impact: how it shook the art world and its consumers. Many people dismissed camera photography as inferior to painting and resisted its integration into the art world as a medium and tool. Today, those individuals are deemed close-minded, as photography has fully assimilated into the art world. At this juncture in history, we might wonder whether those who stand against AI today are the same as the naysayers of the past. Is AI the future? And if so, how will the human art world adapt and distinguish itself? To properly answer these questions, we must examine the impact of photography on the art world and compare its process to AI image generation.
The camera was invented in 1839. Throughout the rest of the nineteenth century, photograph development studios opened and camera technology rapidly expanded. By the start of the twentieth century, almost anyone could capture the world around them in perfect detail within seconds. Expensive portraits that once took hours, if not days to paint, could now be completed in twenty seconds, more accurately, and for a lower price. But not everyone approved of photography. Charles Baudelaire, the famous French art critic and author of Les fleurs du Mal, stated that “the photographic industry was the refuge of every would-be painter, every painter too ill-endowed or too lazy to complete his studies.” Baudelaire was not alone in this sentiment, as many artists and critics dissented the use of photography. Photography was seen as a lazy shortcut, hindering people from pursuing proper artistic education and lowering the standards of artistic accomplishment.
Regardless of public approval, the value of art shifted from the accuracy with which an artist could reproduce their subject to the emotion and perspective their subject could convey (this is not to say that realistic artwork lacked emotion and symbolism). Realistic paintings, while still popular in the nineteenth century, were no longer as impressive compared to the precision of photography, leaving many artists in need of stylistic reinvention. This led to the birth of the impressionist movement in the late nineteenth century. Artists like Manet and Degas were inspired by the ability of cameras to capture individual moments. Their paintings explored light and movement through their vibrant colours. Bold and visible brushstrokes become definitive. Impressionism then gave way to other movements such as post-impressionism, fauvism, abstract expressionism, and many more movements in the wake of the twentieth century. Artists like Norman Rockwell also began to utilize photographs for easy reference to their paintings. Over 150 years after the invention of the camera, photography had fully integrated into the art world as a tool for photo referencing, mixed media work, and as a medium of its own. Much like cameras were similarly denounced, AI art is criticised for being lazy and ‘cheating’ the creative process, removing opportunities from real artists. Many advocates of AI state that AI art is simply the next step of progress in the art world: a tool to streamline creativity.
Returning to our first question regarding those against AI-generated images and whether they differ from yesterday’s camera critics, the answer is no. The surface-level critiques mentioned before of both AI and photography are similar, sure. But concerns about the actual function of these technologies are different. Photography captures the world as it is—its artistry lies in how the subject is captured rather than the subject itself. Additionally, most photographers do not claim to have created what they capture. AI imaging is, however, quite different—and this is the crux of the issue. The products of generative softwares are not sourced from any creative process but rather from scraping billions of images, including real artists’ works, to create what is then labelled an original image. Unlike photography, which exists as an artistic medium independent of the paintings it once competed with, AI generated images cannot exist without the work of the artists that it is replacing. With this, an ethical dilemma arises for artists whose works are scraped by AI. Artists are typically uncompensated and not asked for consent before AI integrates their works into a database. This is especially problematic as freelance and corporate-world artists are replaced by these softwares.
There also comes a comparison of necessary skills required for photography and AI. Where critics of photography went wrong was their statement that photography required no artistic skills. This is not the case at all. Photography requires understanding of composition, colour grading, lighting, and post-processing to be considered successful. AI image generation, however, only requires its users to write a prompt, which the software then creates. Users of these softwares can request changes and possibly learn what words to use to generate their desired images. Knowledge of colour theory, anatomy, composition and other fundamentals of art are not required. If anything, the role of the producers of AI images is more similar to that of a client than an artist, requesting art rather than creating it.
Regardless of opinion, it seems that AI images are here for the long haul, entailing drastic considerations for the art world. Artists cannot adopt new styles as the impressionists did, as these AI models adapt within days, but two methods of artistic protection stand out. The first is through the law. AI, despite its rapid development, is still new, and the law is playing catch-up. As AI software utilises scraped data, much of the images it integrates are copyrighted—the question of whether their use of these materials counts as infringement is currently debated. As more lawsuits are issued against generative AI companies, precedents and legal protections are bound to arise. The second method is through consumers. As consumers, we dictate value. When we consume and interact with AI content, we give it value and contribute to the growth of its industry. If we resist and even vocally reject generated content, it loses its value. Now, more than ever, it is important to be intentional with our consumption. Supporting local artists has become increasingly crucial. While this is, admittedly, easier said than done in our economy, mere interaction with local artists can make a difference.
Despite unprecedented times, hope mustn’t be lost. We are currently in the wild west of AI, but as time passes and restrictions on AI and artistic protections are implemented in conjunction with AI technology’s development (hopefully becoming less burdensome on the planet), AI can serve as a tool for artists much like photography does.