Generative AI and the Commodification of Creativity

I hate generative AI with every fiber of my being, so let’s get that out of the way right now. Does my hatred make this analysis biased? Probably. Does it matter in the long run? Probably not, because the objective evils of gen-AI are pretty overwhelmingly obvious to anyone with more than two brain cells to rub together.

Before we go further, a TW: Violence, self-harm. If either concerns you, skip the next paragraph.

Environmental impacts from massive data centers aside—and those are some tremendously bad impacts—generative AI, by which I mean large language models like ChatGPT or Claude or Grok, are killing creativity. Literally and metaphorically. Literally, because people are actively dying as a result of their “conversations” with these AIs; ChatGPT alone has been directly linked to multiple suicides and homicides through its negligent design. The systems aren’t designed to be used in place of a human therapist, yet that’s how they’re being used. Where a human therapist or chaplain or healthcare practitioner or social worker is a mandatory reporter of imminent violence (i.e., if someone communicates a direct, actionable, believable threat to themselves or others), these systems are designed instead to validate the feelings of their unwitting victims. Instead of saying, “Hey, maybe you shouldn’t do that,” ChatGPT will tell you what it thinks you want to hear. And that has resulted in multiple deaths.

But I am going to take off my law enforcement hat now, since I am no longer involved in that community, having put my military police behind me back in 2021. Instead, I’m going to focus on gen-AI in creative spaces, because that is where its negative side effects are less obvious—but no less insidious. Which brings me to the metaphorical ways gen-AI is killing creativity.

Those who style themselves AI “artists” or “writers” merit neither title. Typing a prompt into an LLM to produce an amalgamation of mediocre slop is not the same thing as devoting weeks, months, years of one’s life into the act of creation. That’s not even counting the years it took to hone one’s craft in the first place. I’ve been telling stories since I could speak. I’ve been drawing since I could hold a crayon. I’ve been writing since I first learned my letters. I wrote and illustrated my first “book” when I was in kindergarten; I actually remember some of it. There was a character named Sean, which I recall informing my mother in no uncertain terms was pronounced “seen” rather than its actual, ya know, pronunciation. Ironically, I later married a man with that name, and my parents found that first book I wrote way back when. Five-year-old me was prescient, apparently.

Life is funny that way, no?

Thirty-odd years, including eleven total years of university (from starting undergrad and through getting my PhD), I’m still improving my craft as a writer, book by book, word by word. AI “writers” are not. They are outsourcing their own ability to think, to improve, to create to a soulless machine, and what are they even producing at the cost of gallons of freshwater and heat and noise pollution? Slop. A sum of averages stolen from the work of human authors. It’s theft, pure and simple—and it’s not even good writing. It is, by definition, average, because it is finding the middle ground among the stolen source material. What we call “AI” isn’t actually an intelligence. Rather, it is advanced predictive text. LLMs produce only what they think looks like a plausible answer. You tell it to write a dark romance novel featuring a given trope, and it’s going to give you something that looks like that, but each word will be chosen not with the care and precision a human author would use, but by choosing whatever word that is most likely to come after the previous one. The word that most often follows the previous word in existing examples of dark romance novels. Hence: average. No gen-AI can achieve greatness; only mediocrity.

But do readers even want greatness anymore? I don’t have an answer for that. Some, surely, still do seek the depth of the human soul in art and literature and music, but the slop oversaturating the market makes truly outstanding creative works harder and harder to find. The pervasiveness of AI slop will only lead to further decline as LLMs cannibalize their own works to produce “new” ones, until every book is just a sum of averages, a copy of a copy of a copy, stolen from itself.

And you can’t even copyright it to begin with, because at its core, it is theft. If our government had any spine, it would make the sale of AI-produced “art” illegal as well. Instead, so-called “artists” and “writers” are profiting off of stolen creativity, at an incredibly steep cost to the environment. Unfortunately, I doubt our government will step up to the plate anytime soon, especially not with silicon valley tech-bros lining congressional pockets.

The brain is a wonderful and complex organ, capable of producing endless creative works. We are limited only by the restrictions of physics and human endurance. Our species has endured this long because we are so adaptable—in other words, so creative—and outsourcing that creativity now will prove a death sentence. Maybe not today, maybe not tomorrow. But eventually, beyond doubt, gen-AI will be our undoing. The brain is a muscle like any other and must be exercised accordingly; we’re already starting to see the result of its atrophy at a societal level. Some people can’t even produce a grocery list or draft an email without consulting ChatGPT, let alone write a novel.

There is no amount of gen-AI in the creative process that is okay. None. Not in brainstorming (and, really, without a brain, can it even be called that?), not in outlining, not in drafting, not in editing. For the record, spell check and basic grammar check, which have been around for decades, are not in the same realm; I’m talking about actually plugging a written work into an AI model to replace a human editor. In accepting the prevalence of AI slop as inevitable, we have turned the act of creation itself into something that can be bought and sold, allowing anyone to style themselves as an artist or writer or musician, when they are doing nothing more than typing a prompt into a machine.

At the end of the day, not everyone should write a novel. Not everyone has the soul of a poet, the heart of an artist, the callused fingertips of a musician. Technique can be taught, but technology cannot replace talent. Until we demand better accountability, we’re doomed to keep turning Idiocracy from satire to reality.