Using AI to write your story

Posted on November 7, 2025 by Michael A. Ventrella

Don’t. Dear whatever god(s) you believe in—just don’t.

Artificial intelligence can do some good things, but being creative isn’t one of them. If you use AI to write your story, you’re stealing other people’s creativity, and you’re not going to get a very good result.

And editors can tell. It’s not well-written. It’s full of cliches, has no original ideas, and there’s no poetry in the words.

Is that what you want? Predictability? No surprises in your work? Something so basic that even a computer could write it in your sleep? Publishing that would make you proud of your accomplishment?

AI can be useful for research, and, let’s face it, it’s not going away. It’s going to get better and more reliable—but we’re not there yet. Remember, all AI does is grab stuff from the internet, and the internet is full of crap. AI doesn’t know if something is “fake news” but hey, it may help you get started in your research as long as you realize what it gives you may be completely wrong.

(As an aside, I gave Google’s AI the prompt “Why you should not use AI to write your stories” and here’s what I got: “Unlike you—an author—AI won’t produce an original idea, a compelling plot twist, or a unique structure. If you rely on AI to write your book—or significant portions of it—that content will likely be predictable, generic, and, frankly, boring.” I agree with that predictable, generic, and frankly boring response.)

AI is based on statistics, not creativity. It predicts what the next most likely word will be (like your phone predicts your texts) and carries that to the extreme, where it can predict what the most likely plotline will be.

Sadly, some people think that AI is an easy way to get rich quick. You pop in an idea, and the computer spits out a story for you that you can self-publish (because no legitimate publisher will take it) and soon, people will buy it! How easy! The people who do that have no desire to be writers—they just think it’s a way to make money.

And then the stuff doesn’t sell, because it’s crap. Oh, it may make a few dollars, but no one is going on lavish vacations with the money earned from AI stories.

So what about using it as a tool, as a way to figure out a plotline? Once more, you’re relying on a system that has no creativity of its own. All it can do is steal the creativity of others. It might give you an idea for resolving a plotline that has been done a hundred times before, and even if you don’t recognize where the idea came from, some reader somewhere will.

I like to talk about stuff I’m writing with creative people and not robots, because sometimes, by talking it out, I can figure out a major plot point or otherwise work out a problem I was having with the story. Talking to boring people doesn’t help me at all. Talking to people who have no creative spark won’t give me what I need. And AI has no creative spark.

A lot of what editors find in AI submissions are stories that don’t make sense. The AI doesn’t know right from wrong, it doesn’t know what works, it doesn’t understand pacing and foreshadowing and the techniques writers use to make the story come alive. It just spits out stuff based on what it steals on the internet, and like AI artwork, sometimes it has too many fingers. Neil Clarke, award-winning editor of Clarkesworld magazine, says it this way: “AI can’t do subtext. It does very flat stories. Sometimes it will miss a plot point. It will hop over something that would be obvious, but it doesn’t know, because it doesn’t know anything.”

Is it possible that you might submit a story and have the editor reject it because the editor thinks it’s an AI story when it really isn’t? Sure, that’s happened. But what does that tell you? It tells you that your story is predictable, boring, and unimaginative, like an AI story. Keep working at it. You’ll get better.

It all boils down to what someone said about AI: “Why should I spend time reading a story no one has spent time writing?”