AI art is a sign of the times. Automation is at the forefront of our world, soaring past its initial introduction on the factory floor in 1785 and evolving into something intelligent and capable of high-level decision-making. Its introduction into the art world is somewhat surprising, considering the widely accepted standard of art as the expression of human creativity and imagination. While art has certainly evolved over the last centuries, it has existed within a paradigm of human emotion and creativity. 

Sites like DALL-E — developed by OpenAI and Midjourney — allow a user to enter a keyphrase and receive a custom digital image. DALL-E, named after surrealist painter Salvador Dalí and Pixar’s futuristic movie “WALL-E,” allows fifteen free credits per month with the option to purchase more. A Shiba Inu dog wearing a beret and a black turtleneck, an Andy Warhol style painting of a French bulldog wearing sunglasses and a photo of a teddy bear on a skateboard in Times Square are just some of the examples one sees traversing the home page.  

Many deem AI art generators powerful for their ability to allow anyone to create art. However, one could also argue that anyone can create art without AI, especially given the past century’s move towards increasingly abstract forms of art and away from previously dominant discourses around correct proportion and preference for trained artists. Regardless of what’s accepted in the art world, AI certainly removes some of the pressure and shame that some people may feel about creating art. Even for trained artists, AI can assist in the conceptualization process, allowing a greater connection between the artist’s mind and what they put on paper.

Creation is not the only way AI art has made visual art more accessible. It has also changed the way we view and engage with art. These services are free to use, allow for connection with the work and show users the possibilities that art brings. Museums and special exhibitions can be costly, and purchasing art is much more so. These services remove these costs and add a sense of interaction to the art viewing process. Nonetheless, interaction with AI art can also stem from inspiration from a museum or gallery experience. It allows a user to see how older artists might interact with modern themes. One example painting that occupies the homepage of DALL-E is entitled “an oil painting by Matisse of a humanoid robot playing chess.” Beautifully rendered in Matise’s strong colors and fierce brushwork, the work elicits themes of automation and technology. The work references online chess play’s removal of previous forms of human interaction and conversation that exist in parks and public spaces. It hosts a warm orange backdrop but is countered by the cold faceless nature of the robot sitting across the table. 

The rendering of this striking, Matisse-inspired piece raises questions about copyright and plagiarism. Generating “avocados by Barbara Kruger” produces interesting results accurately crafted in Kruger’s collage caption style. An AI generator scours the internet for existing images, video, text and other media and quickly feeds it into the generator, all without regard for trademark or requesting the use of artists’ materials. This results in modern artists losing money to the cheaper alternative of traversing an artist’s unique style on a customizable basis for free. One could imagine a young artist spotting their style on an AI art bot thread, cheapening their skills and product. Artists are left uncompensated for AI art. This issue has manifested itself quite publicly in the art world. The 2022 Colorado State Fair art competition first place recipient in the digital art category was Jason M. Allen, who used Midjourney, an AI generator, to create his piece Théâtre D’opéra Spatial, causing outrage in the art community.

Unfortunately, copyright laws are relatively obsolete against the power of AI generators. AI has grown faster than lawmakers can legislate, creating a complicated time in the history of copyright law. AI is still at the early stages of its development, and the best way to regulate it is still unclear. AI image generation can also cause reputational damage and be used to spread misinformation. The term “deepfakes” has been coined to describe a type of synthetic media in which a person in an existing image or video is replaced with someone else. Experts indicate that AI image generation will quickly evolve into video generation, furthering concerns that this type of technology may erode trust in institutions. 

While generally considered a modern phenomenon associated with AI, photo doctoring has long existed as a method of manipulating the public. After consolidating his power in 1929, Josef Stalin declared war against opposing Soviets and attempted to rewrite history through censorship and erasure of older Soviet officials from photographs. Despite the deeply problematic history of photo doctoring, the new widespread accessibility of high-level deepfake technology is perhaps even more concerning. This reality, combined with the rampant textual misinformation that traverses social media, is yet another aspect of the current crisis that exists in news and media. Transparency, security and copyright are important next steps for AI art as algorithms are scrutinized for unpredictability and use of personal images. 

AI, while still in its early stages, is quickly imposing its influence on the world in a wide array of areas. This quickly emerging technology has greatly expanded the public’s access to art, both on the creation and the engagement side. However, the lack of regulation around the technology and the possibility of misuse is concerning. AI art represents a microcosm of a broader question of how we adjust to new technology, how we can use it ethically and how we harness it to enhance the human experience rather than harm it.