....and that, not the technology, is what should scare us.
Artists are protesting, lawsuits are piling up, and “AI slop” has become a standard insult. But I think much of this resistance is aimed at the wrong thing, because it treats the technology itself as the threat, as if the problem were that machines are entering a space that belongs to humans. The truth is, art has never been safe from new technology, and it has never needed to be. Art has always reacted to the world around it, and it has always been changed by social transformation, technological progress and the breaking up of old patterns. New technologies have allowed artists new forms of expression, and the history of resistance against them is long. Photography (for example) was once dismissed as mechanical, soulless and unworthy of the name art, and it turned out to be one of the best things that ever happened to it.
These technologies did not only change how art looked, they changed who was allowed to take part. Photography allowed art to become a mirror of society as a whole and turned it into something societal. It brought ordinary lives, overlooked people and uncomfortable realities into the picture and made art a shared space. It made it possible for directions to emerge that would otherwise have had no chance, because they were considered too “vulgar” or too “political.” Democratization brought diversity, and diversity cracked open the narrow view of an elite that had owned the picture of the world for far too long.
Every major technology before AI pushed in that same direction, for one reason: each was a tool, and every tool still needed a person. Someone with a life, a viewpoint and something to say, standing in a particular place, telling their story and showing emotions. The technology widened who could act but the act itself was human!
This is where generative AI breaks the pattern, and the danger lies not in the machine’s existence but in what it does to authorship. Its defining feature is that it can produce complex, varied creative output from a single high-level instruction, with the human contributing almost none of the execution. Yes, you can argue that prompting, selecting and iterating are creative choices, and to a degree they are. But they sit at a completely different distance from the work than composing, framing or painting. Photography as a technology democratized participation: more people could make art out of their own experience. AI democratizes output: more images exist, while the person’s creative act shrinks to a request. Those are not the same thing, and confusing them is how “anyone can make an image now” gets mistaken for democratization.
Once the act can be bypassed, art splits in two. On one side, a loud, fast, commercial sector where AI images pour out cheaply and endlessly. On the other, a high-art sector that comes back stronger than ever, where the imperfect and the human are treasured as exclusive. What isn’t democratized rises in value again. What is human made becomes a luxury. And the elite that the democratization of art once pushed aside returns, not by locking people out of the technology, but by owning the one thing the technology can’t supply.
This is uncharted territory, and it forces a question we have never had to face. The boundary between human expression and machine output has never been this blurred. When art becomes mass output, devalued to near zero, value can no longer live in the output itself. So where does it live then? As long as only humans could make art, we never had to say. Now we do. If the value is no longer in the output, it has to be in what the output can’t fake: the vulnerability, the lived experience, the real risk, the personal conviction someone pours into a work. I don’t have the full answer, and I suspect none of us does. But if we can’t say what a human adds, the market will decide for us, and it has never been sentimental. The machine will never ask what art is for. We have to, before it stops mattering.
