Artificial Expressionism

Each of these works starts with a text prompt (often timeless issues like floods, fires, protests or art history references). Then a conversation happens—a back and forth—the text is tweaked, images are selected, sections refined, variations generated, until the composition is set initially, and further refined as they are executed as a painting on canvas (usually in acrylic).

In this way they are a dialogue between me and an artificial intelligence (in most cases DALL-E by OpenAI). Although, artificial intelligence might be a bit strong, the program is really a neural network, or maybe think of it as consensus engine. It (they?) “reads” the Internet, models that information, builds correlations, and creates weightings and associations. In this sense it builds a consensus—not a Platonic ideal but more of an Internet ideal—the sum thinking of images available on the Internet. It thus produces an average, and for many things the output is just that, average.

Computers are useless. They can only give you answers.

— Pablo Picasso

But the corner cases are interesting—the odd combination of concepts, objects, and styles. There are times when things don’t add up. The program has no sense of space, or what physical reality requires and doesn’t. It doesn’t know objects or physical space in the traditional sense—only weightings and correlations—it doesn’t know our rules. Let alone what it thinks dread or despair looks like. And in this way, you can coax out interesting scenes, situations, and even aesthetics. It is where things diverge that the images become interesting—twisted faces, illogical rooms, and the odd combinations or insertions.  There are splatters and scrapes, zips and color-fields, and different textures and tones.

So the genesis of these works are artificial—the initial composition comes form a neural network trained on the Internet. However, what they are trained on is made by humans (as least for now) and some of the images are deeply human and here lies the dichotomy. The output is human, expressive, interesting, curious, haunting, and maybe even true despite the artificialness of how they were created, a form of artificial expressionism if you will. That’s the interesting thing for me, the tension—the not quite right and the humanism in many of the images.

So maybe now computers can give us questions and not just answers.