AI Writers Retreat  ·  AI & The Craft  ·  Issue No. 4
AI & the Craft — A Series for Writers

Don't Let It
Sand You Down.

AI has read more writing than any human alive. That makes it powerful. It also makes it have very strong opinions about what writing is supposed to sound like — and those opinions skew toward the average.

AI, Voice & The Flattening Problem February 26, 2026

There is a particular kind of workshop feedback that every writer has received at least once. It arrives in a reasonable tone, from a reasonable person, and it is completely technically correct. The sentence it flags is grammatically unusual. The rhythm it questions is deliberately broken. The word it suggests replacing is, objectively speaking, a perfectly good synonym for the one you chose. And yet, if you followed every note, the piece would be worse. Not different — worse. Smoother, safer, less itself.

AI gives this feedback constantly. Not because it is wrong, exactly. Because it is right in the way that averages are right. It has been trained on an ocean of human writing and has learned, with frightening precision, what most good writing looks and sounds like. The problem is that your writing — if it has a voice worth keeping — is almost certainly not most writing.

This is what we mean by the flattening problem. It is not that AI will ruin your voice in one dramatic intervention. It is that it will nudge, gently and persistently, toward the mean. And if you're not paying attention to which notes to take and which to refuse, you can revise your way into something technically polished and utterly anonymous.

The most dangerous AI feedback is not the note that is wrong. It is the note that is correct about everything except what makes your work yours.

Why the Average Is So Seductive

The feedback that flattens doesn't feel like flattening when it arrives. It feels like an improvement. That's what makes it worth understanding.

Consider a writer whose voice is built on long, winding, clause-stacked sentences — sentences that accumulate like weather, that make the reader lean in before they know why. An AI trained on writing advice, style guides, and workshop feedback will flag these sentences every time. Too long. Consider breaking this up. Readers may lose the thread. The note is not wrong. Most readers do prefer shorter sentences. Most writing does benefit from tighter syntax. But most writing is not this writer.

The same is true of unconventional punctuation, deliberate repetition, abrupt tonal shifts, fragments used for rhythm, the refusal to resolve what the reader expects to be resolved. All of these are potential fingerprints — the things that make a voice a voice — and all of them will trip the AI's pattern-matching in the direction of "this could be cleaner."

The writer's own sentence
"She had been waiting for him to say something worth hearing for so long that when he finally did — finally, after all of it, after the years and the silences and the particular cruelty of his kindness — she almost didn't recognize it as the thing she'd been waiting for."
Intentionally long. Clause-heavy. The accumulation is the meaning — the reader experiences the wait.
After "helpful" revision toward clarity
"She had waited a long time for him to say something meaningful. When he finally did, after years of silence and complicated kindness, she almost didn't recognize it."
Shorter. Cleaner. Technically correct. The wait is described. It is no longer felt.
What the revision removed
The repetition of "finally." The parenthetical pile-up that enacts the exhaustion. "The particular cruelty of his kindness" — a line that earns its length. The fragment-like trailing clause that keeps the sentence from resolving. The sentence was the experience. The revision is a report about it.
The flattening was not accidental. Every change was individually defensible. Together they removed the writer.

What AI Is Actually Responding To

Remember what we covered in Issues 1 and 2: the AI processes tokens, and it predicts what comes next based on statistical patterns in everything it's been trained on. This means that when it reads your deliberate fragment, it is not asking whether the fragment is intentional. It is asking whether, in the vast landscape of writing it has seen, a fragment in this position is more likely to be intentional or a mistake. And the honest answer, statistically, is: usually a mistake.

The model isn't wrong to flag it. It just cannot, on its own, distinguish between an error and a choice. That distinction requires something the AI doesn't have access to: your intention, your awareness of the pattern, and your knowledge of whether this is the third time you've used this device or the first.

What AI Sees

A sentence fragment where a complete sentence is statistically expected. A word repeated where variety is the norm. A comma splice where most style guides say no. A tonal shift where consistency is more common.

What You Know

That the fragment is doing the emotional work of a full stop. That the repetition is a drumbeat. That the splice is the breathless pace of panic. That the tonal shift is the point. The AI is missing the intention. Only you have it.

How to Protect Your Fingerprint

This is not an argument against using AI for revision — it is an argument for using it with your eyes open. The flattening problem is real but entirely manageable, once you know it's there. Here are the practices that keep your voice intact.

1
Name Your Devices Before You Ask
+
Before handing your draft over for revision feedback, tell the AI explicitly which elements are intentional. This is a direct application of context engineering from Issue 3 — you are briefing your reader before they read. When the AI knows that the long sentences are a feature, not a bug, it stops flagging them as problems and starts reading them as choices. The feedback shifts from "fix this" to "does this work?"
"Note before you read: the run-on sentences are intentional — they mirror the narrator's anxiety. The repetition of 'still' is a deliberate motif. Do not suggest I break up the syntax. Read for whether the accumulation is earning its length, not whether it conforms to standard style."
2
Ask It to Read for Effect, Not Correctness
+
The default mode of AI feedback is editorial — it looks for what can be improved against a standard. Shift the frame. Instead of asking what's wrong with the prose, ask what the prose is doing. What effect does this sentence have? What does the rhythm communicate? Does the unconventional punctuation land as intended? These questions invite the AI to respond to your work on its own terms, not against a generic template.
"Don't evaluate this passage against standard style. Tell me what effect the syntax is having on you as a reader. Does the length feel earned or exhausting? Does the repetition feel deliberate or accidental? I want to know if the device is working — not whether it's conventional."
3
Know Your Own Fingerprints First
+
You can only protect your voice in revision if you know what it actually consists of. This sounds obvious and is surprisingly hard. Most writers have a felt sense of their voice without a precise vocabulary for it. Before using AI for revision on work you care about, spend time naming your own devices: the sentence lengths you favor, the punctuation habits, the structural patterns, the tonal registers you move between. When you can name them, you can defend them. When you can only feel them, the "helpful" note will always sound reasonable.
Keep a voice inventory: 3–5 sentences from your best work that are distinctly yours. Return to them before revision sessions. Use them as a tuning fork. If the revised draft no longer sounds like those sentences, something has been lost.
4
Treat Stylistic Notes as Questions, Not Verdicts
+
When AI flags something in your prose, the most useful response is not to accept or reject — it's to ask why. If it suggests cutting a sentence, ask what the sentence is currently doing that makes it feel unnecessary. If it recommends a synonym, ask what the original word is conveying that prompted the suggestion. The note becomes a flashlight rather than a red pen. You use it to look more closely at your own choice, and then you decide whether the choice is doing its job.
"You flagged the sentence beginning 'And yet.' Rather than changing it, tell me what that sentence is currently doing in context — what job it's performing, and whether it's succeeding. I'll decide whether to revise from there."
The goal is not to ignore AI feedback about your style. The goal is to make sure you are choosing — not just accepting. Every note you follow should be a decision, not a default.

The Distinction That Protects Everything

Underneath all of this is a single distinction worth carrying into every revision session: the difference between a mistake and a choice.

A mistake is a place where the writing failed to do what you intended. A choice is a place where the writing is doing exactly what you intended, and the question is only whether it's succeeding. These require completely different responses. Mistakes should be fixed. Choices should be interrogated — pushed on, defended, occasionally abandoned — but always from a position of awareness.

AI feedback cannot make this distinction for you. It doesn't know what you intended. It only knows what landed. Which means that when you bring your work to an AI for revision, the most important thing you can do is know, before you start, which category each element of your prose falls into. Armed with that knowledge, the feedback becomes genuinely useful — a skilled outside eye on the places where your choices aren't yet succeeding, and a prompt to look harder at the places where they are.

The flattening problem is real. But it only has power over writers who have forgotten the difference between what they chose and what happened by accident. Know your own work that well, and no amount of helpful suggestions can sand you down.

AI cannot tell the difference between a mistake and a choice. That distinction is yours to keep.

END OF ISSUE NO. 4 — AI & THE CRAFT