Two sentences. That's all it takes now. The prose has a faint plastic sheen, the paragraph lands every point and means nothing, and some part of your reader-brain has already filed it under machine and started skimming. You didn't decide to do that. The instinct just fired. And it's getting sharper by the month.
That's the whole problem for anyone trying to use these tools honestly. The goal isn't to trick anyone. It's the opposite — AI writing that doesn't sound like AI is what you get when the human does the parts that matter and the machine does the grunt work. Two different jobs. Most people have them backwards.
Let me name the tells first. You can't fix what you can't hear.
How readers catch it
The default voice of a language model is a specific, recognizable thing. Once you've clocked it, you see it everywhere — cover letters, newsletters, that suspiciously even-tempered Amazon review. Here's what gives it away:
- Over-polished symmetry. Every sentence the same medium length. Every paragraph three sentences. Clauses balanced like a furniture showroom. Real writing limps and sprints. It has a bad knee.
- Hedging on everything. "Can be," "often," "may help," "in many cases." The model is trained to never be wrong, so it commits to nothing. A human writer plants a flag.
- The summary reflex. It restates what it just said, then tells you what it all means, then ties a bow on it. Nothing is allowed to just sit there and be unsettling.
- No real specifics. It'll say "a delicious meal" where you'd say "cold pad thai eaten over the sink at 1 a.m." The model reaches for the category. A writer reaches for the thing.
- No opinion. It describes the controversy from a respectful distance. It will not tell you the second act drags, or that the villain is more interesting than the hero. It has no taste, so it pretends taste doesn't exist.
- The same three-beat rhythm. Setup, elaboration, gentle conclusion. Over and over, like a song stuck on one chord.
None of these is really about word choice. They're about stance. The model has no body, no grudges, no Tuesday it can't stop replaying. So it writes from nowhere, about everything, to no one in particular. That's the smell.
Write the messy human version first
Here's the single biggest shift, and it's about workflow more than prompting: don't let AI originate. Let it expand.
Ask a model for a scene from a blank prompt and you get the average of ten thousand scenes like it. The orphan is plucky. The rain is symbolic. The mentor dies in act two. A model can only hand you the center of the distribution, and the center is exactly where everyone else already is.
So write the ugly version yourself first. Not a polished draft — a beat. A few rough lines carrying the things only you know:
She apologizes by overwatering his plants. He notices the dripping pot, says nothing, gets it.
Barely a sentence, and already more alive than anything a cold prompt produces, because it runs on a specific, slightly off emotional logic — her logic. Now AI can expand it. Give you the staging, the connective tissue, three versions where the pot drips onto a stack of his unopened mail. You stay the author of the idea. The machine upholsters it.
This is why a beat-at-a-time workflow beats one big "write me a chapter" prompt. You feed the model your raw material instead of asking it to invent yours. When you draft rough beats and then expand, rewrite, and split them one at a time, you're the one deciding what the scene is about before a single word gets generated. The AI never drives. It navigates the parking lot you built.
Rewrite line by line, never accept the paragraph whole
The lazy move: generate a paragraph, skim it, decide it's "good enough," paste it in. Do that across a novel and you get a book that's 90 percent good-enough. Good-enough is the exact texture readers have learned to distrust.
Better: treat every generated line as a suggestion you're arguing with. Read it out loud. Where does it go slack? Where did it reach for the obvious word? Kill the hedges. Cut the sentence that summarizes the previous three. Swap one generic noun for the real one only you would know to use. You'll often keep the model's structure — the order of images, the rhythm of an exchange — while replacing half the actual words. Fine. That's the work.
A loop that keeps you honest:
- Generate two or three versions of a beat, not one.
- Steal the best line from each. Discard the rest without guilt.
- Rewrite the stitched result by hand until it sounds like you on a slightly better day.
The point is that the final sentences passed through your judgment, one at a time. A reader feels that even when they can't say why.
Feed it your voice, and the specifics it can't invent
A model defaults to a voice — competent, neutral, faintly corporate. If you want yours, you have to show it. Paste in a few paragraphs of your best prose and tell it to match the cadence, the sentence shapes, the way you handle dialogue. Keep a "voice sample" somewhere you can reuse. A good story bible — the context notes the AI reads before it writes — should hold more than plot and character facts. It should hold a sample of how the book actually sounds. Then the expansions come back closer to your wavelength than to the default one.
And guard your specifics like the last matches in a cold cabin, because they're the one thing the machine genuinely cannot produce:
- The exact wrong thing your character says at the funeral.
- The smell of the apartment that tells us the marriage is over before anyone speaks.
- The regional word. The inside joke. The texture of a real job.
- An opinion the narrative actually holds — about the characters, about the world, about what matters.
These are non-negotiable. The model will paper over a missing specific with a pleasant generality every single time, and that generality is precisely the AI smell. So you keep a list. You insist. When a generated passage gives you "she felt overwhelmed," you delete it and write what being overwhelmed looked like for this person, in this kitchen, on this day.
Do the hard human pass at the end
However careful you were beat to beat, read the whole thing through once more with a red pen and a foul mood. This is where you catch the drift the line-level edits missed: the chapter gone suspiciously smooth, the three paragraphs that all open the same way, the opinion that quietly leaked out and left everything sounding fair and dead.
Hunt for the tells from the top of this piece. Search your own draft for "began to," "a sense of," "it was clear that," "couldn't help but" — the connective tissue models over-produce. Read dialogue aloud. AI dialogue is too articulate, too on-topic; nobody interrupts or misunderstands. Add the friction back. Let someone answer the wrong question.
If you're stitching together a long project this way, the editing discipline matters more, and it scales — the same instincts that fix one paragraph keep a whole manuscript sounding human across an entire novel. And on the days the blank beat itself won't come, that's a different problem with its own fixes for when you're stuck.
The throughline
Think of the model as a very fast, very well-read intern with no taste and no skin in the game. A power tool. A table saw doesn't design the cabinet — it cuts faster and straighter than you can by hand, and it'll happily cut the wrong board if you let it.
So keep your hands on it. Decide what every scene is about. Write the strange specific seed, let the machine expand it, then rewrite the expansion until it bleeds your voice and not its own. Work a beat at a time. Stay in the driver's seat. Do the human pass no amount of compute will do for you.
Get it right and nobody ever asks whether AI touched your book. They're too busy reading it.