On AI-Generated Imagery

Why computers cannot and will never be able to truly create art.

I had hoped that the entire AI fad would have run its course in the public eye by now, but considering both private industries and the White House have continued to pump billions into that whirlpool of excess fingers, I thought it necessary to explain why the images they call art aren’t art at all. There is no trick of rhetoric here to convince people of this position, but instead we must apply widely-accepted definitions to these situations to rise to a higher truth. Under a more forgiving timetable, this discussion may benefit from the flourishes that a Socratic dialogue would produce, but I’ll keep that in my back pocket for another day.

According to Euclidean geometry, it is axiomatic that the whole is equal to the sum of all its parts. The universe, of course, cannot be measured entirely using Euclidean geometry. “Art” is the word we use to describe whatever falls outside of that boundary. The cost for a painter to paint a room in a house is the sum of the cost of the materials, the size of the area, the time it takes to do the labor, and some additional surcharge for his expertise to make a profit. A muralist may take longer to paint an area of a comparable size, but this does not account for the exponential cost of the project, nor does it account for the astronomical monetary value it adds to the building in the right hands. Indeed, graffiti artists like Banksy use stencils to reduce the time it takes for them to paint on their medium to an absolute minimum. Emotional value accounts for these augmented factors.