AI Can Generate Images, But It Can’t Make Art: Why Human Judgment Always Wins

Generative AI can produce an image in seconds. It can mimic styles, combine references, and generate visual output at a scale no human can match on speed alone.

It cannot decide what to draw. It needs a human to prompt it.

It cannot look at a written scene and translate it into a sequence of panels that control pacing, emotion, and the reader’s eye movement across a comic book page. It cannot choose the camera angle that makes a moment dramatic or poetic. It cannot use negative space as a storytelling device. It cannot place a figure in the lower third of a panel because the emptiness above them is the point or waiting for lettering.

These are not technical skills. They are judgment calls. And judgment is the product of decades of drawing, reading, observing, failing, and drawing again.

I've watched this debate closely for the last two years. I've written about the legal cases (Bartz v. Anthropic, Thomson Reuters v. Ross, the Disney/Midjourney fallout). The law is slowly catching up to the technology. But the more important argument is not legal. It's creative.

The value of an illustrator was never the labor of moving a tool across paper, canvas, or a screen. The value was always in knowing where to put the lines and why. That is not something a prompt can do. Not because the technology isn't good enough yet, but because it requires a human being who has lived inside the creative world long enough to make decisions that a model trained on scraped images simply cannot mimic.

I'm not betting on the tool. I'm believing in the artist who knows how to use it.

Just create.™