In just one few years, the number of artworks produced by self-described AI artists has grown dramatically. Some of these works have been sold by major auction houses for staggering prices and have found their way into famous curated collections. Originally spearheaded by a few tech-savvy artists who adopted computer programming as part of their creative process, AI art has recently been embraced by the masses as image-making technology has become more efficient and easier to use without coding skills.
The AI movement is riding on the cusp of technical progress in computer vision, a research field dedicated to designing algorithms that can process meaningful visual information. A subclass of computer vision algorithms, called genetic models, takes center stage in this story. Generative models are artificial neural networks that can be “trained” on large datasets containing millions of images and learn to encode their statistically significant features. After training, they can produce entirely new images not contained in the original dataset, often guided by text messages that explicitly describe the desired results. Until recently, the images produced through this approach remained somewhat lacking in coherence or detail, although they possessed an undeniable surrealist charm that attracted the attention of many serious artists. However, earlier this year the Open AI technology company unveiled a new model – dubbed DALL·E 2 – that can create remarkably consistent and relevant images from almost any text message. DALL·E 2 can even produce images in specific styles and imitate famous artists rather convincingly, as long as the desired effect is sufficiently specified in the prompt. A similar tool has been released to the public for free under the name Craiyon (formerly “DALL·E mini”).
The coming of age of AI art raises a number of interesting questions, some of which—such as whether AI art is really art, and if so, to what extent it is really made by AI—are not particularly novel. These questions echo similar concerns once raised by the invention of photography. By simply pressing a button on a camera, someone with no painting skills could suddenly capture a realistic depiction of a scene. Today, a person can push a virtual button to run a production model and produce images of almost any scene in any style. But cameras and algorithms don’t make art. People do. AI art is art, made by human artists who use algorithms as yet another tool in their creative arsenal. While both technologies have lowered the barrier to entry for artistic creation—which calls for celebration rather than apprehension—the amount of skill, talent, and intent involved in creating interesting works of art should not be underestimated.
Like any new tool, generative models introduce significant changes to the art-making process. In particular, AI art expands the multifaceted concept of curation and continues to blur the line between curation and creation.
There are at least three ways in which making art with artificial intelligence can involve curatorial acts. The first, and less original, has to do with the curation of expenses. Any production algorithm can produce an indeterminate number of images, but not all of them will be rendered in standard art condition. The process of curating results is well known to photographers, some of whom routinely capture hundreds or thousands of shots from which few, if any, may be carefully selected for viewing. Unlike painters and sculptors, photographers and AI artists have to deal with a multitude of (digital) objects, the curation of which is an integral part of the artistic process. In AI research in general, the act of “selecting” particularly good results is seen as bad scientific practice, a way to misleadingly inflate the perceived performance of a model. When it comes to AI art, however, cherry-picking may be the name of the game. The artist’s intentions and artistic sensibility can be expressed in the very act of promoting specific results to the status of works of art.
Second, editing can also be done before any images are created. In fact, while “curation” as applied to art generally refers to the process of selecting existing work for display, curation in AI research colloquially refers to the work involved in creating a data set to train an artificial neural network. This work is crucial because if a dataset is poorly designed, the network will often not learn how to represent the desired features and perform adequately. Moreover, if a data set is biased, the network will tend to reproduce, or even reinforce, such bias—including, for example, harmful stereotypes. As the saying goes, “garbage in, garbage out”. The saying applies to artificial intelligence art as well, except that “garbage” takes on an aesthetic (and subjective) dimension.