Why ChatGPT rewrites the whole document when you only asked for one fix
You spent thirty-one minutes on a draft. You asked ChatGPT to tighten the third bullet under "Architecture." It returned a different document. Different headings, different examples, a paragraph about "delight velocity" you have never approved of, and — somewhere in there — a bullet that is still wrong.
Reader, it is not your prompt.
The chat box has no way to express target selection. When you describe a passage in prose — "the third bullet, not that one, the other one" — the model has to infer which passage you mean, and on a long answer it plays it safe by regenerating the neighborhood. The fix isn't a better prompt. It's a better format: paste the exact quoted passage alongside your note, separate blocks with ---, and prefix with "apply these edits; leave everything else unchanged." The model stops guessing. Full walkthrough in the guide; the tool is OMGfixMD.
It isn't that ChatGPT is bad at following instructions
The instinctive theory, after this happens to you the fourth time, is that the model is being lazy — that it is easier to regenerate a paragraph than to surgically edit one, so it regenerates. This is half right. It is easier. But the model is not lazy; it is coordinate-blind.
When you type "fix the third bullet under Architecture," the model receives a string of tokens and has to map the string "the third bullet under Architecture" to a specific span of its previous output. On a short answer, this mapping is cheap and usually correct. On an 1,800-word answer — one with two "Architecture" sections, four lists, and a nested bullet that could plausibly count as "third" depending on how you count — the mapping is expensive and often wrong. The model's fallback under ambiguity is not to ask. It is to regenerate widely enough that something in its output will look like the fix you asked for, and the rest of the document gets dragged along for consistency.
This is true of Claude, ChatGPT, Gemini, and Cursor. It is true of every frontier model. It will probably be true of the next three generations of them. The root cause is not model capability — it is that prose descriptions of a passage are not a reliable addressing scheme, and the chat box offers no alternative.
The four things you've already tried
"Only edit the third bullet under Architecture."
You add "only" to the prompt. The model says "Of course! Here is the edited document." and returns a document with two rewritten sections, three reordered paragraphs, and — somewhere in there — a bullet that is now shorter but also mentions "stakeholder alignment," a phrase you had specifically removed on the prior turn.
"Leave everything else unchanged."
You add the magic sentence. It works about 60% of the time on a short answer, 30% of the time on a long one. When it doesn't work, the model will sometimes tell you "I have preserved the rest of the document," right above a document whose opening paragraph has been silently rephrased.
"Do not regenerate the whole thing."
A negative instruction. Language models process negatives, but they are weaker signals than positive targets. "Do not rewrite X" without a precise definition of X is an invitation for the model to decide what counts. It will decide generously.
The ritual: "Here is the full document. Apply only the following change…"
You paste the document back in. You paste the change you want. You frame it carefully. This moves the odds from 30% to maybe 75%. A real improvement. Still not reliable. Still about to fail the next time you need to ship at 11:48 PM, which is the moment you needed it to work.
All four attempts share one assumption: that you can fix this with better prompting. You cannot. Prompting-as-a-lever has a ceiling, and on this problem the ceiling is below reliable. The ceiling is where it is because the underlying channel — free-form prose describing a span of text — is the wrong shape for the job.
What's actually happening
Two things are at work, and it's worth keeping them separate.
First, attention is not a laser pointer. When you tell a model "fix bullet three," that instruction tokenizes into a few dozen tokens, each of which competes for attention against every token in the 1,800-word document you're asking it to edit. The document wins, because the document is larger and more locally coherent. The model then tries to satisfy your instruction inside the context of the document, and it has to guess which span "bullet three" refers to.
Second — and this is the bigger driver — RLHF trained the model to return documents that look complete. During post-training, outputs that returned a clean, coherent, whole document were preferred over outputs that returned a surgical one-line change that left neighbors rough. The model internalized "when in doubt, hand back something that reads like a finished draft." On a targeted-edit request with a prose description of the target, "in doubt" is the default state. So the model regenerates adjacent passages for stylistic consistency — not because attention forced it to, but because its training rewards it for doing so. (OpenAI's published Model Spec and Anthropic's prompt-engineering guide both describe variants of this preference; it's an intentional behavior, not a bug.)
None of this is malice. None of it is bad training. It is the shape of the problem. The model is doing exactly what the chat box asked it to do — produce a coherent next turn — and the chat box did not ask for targeted edit only, because the chat box has no way to ask for that.
What the failure actually looks like
Here is the shape of a real turn, condensed. You paste the draft (let's say a 400-word product brief) and type:
Only fix the third bullet under Architecture —
it should say "stateless" not "stateful".
Leave everything else alone.The model returns the whole brief. The third bullet under Architecture now says "stateless" (good). The opening paragraph has been rephrased from four sentences to three (nobody asked). The bullet above the one you meant is now worded differently (nobody asked). A new paragraph on "scalability considerations" has appeared between the headings (strongly nobody asked). Net result: you got the edit you wanted plus three edits you didn't ask for and one paragraph you will have to delete.
Now the same request in paired-passage form:
Apply this edit to your last answer. Leave everything else unchanged.
"The service is stateful across requests."
[Factually wrong] Rewrite as "The service is stateless across requests."The model returns: the edited document, with exactly one change — that one sentence. Opening paragraph untouched. Adjacent bullets untouched. No new "scalability considerations" paragraph. This is the same model, on the same draft, on the same turn number. The only difference is the format of the feedback.
The format that does work
The reliable move is to stop describing the passage and start quoting it. Paste the exact text of the passage — verbatim, punctuation and all — followed by your note. Separate each such block with ---. Prefix the whole thing with an instruction to apply the edits and leave everything else unchanged. Here is what the model actually sees:
Apply these edits to your last answer. Leave everything else unchanged.
---
"The system leverages a cross-functional synergy touchpoint"
[Off tone] Too corporate. "connects to X" is plenty.
---
"moreover,"
[Delete] One "moreover" per page. This is the third.
---
"delight velocity"
[Delete] We do not measure this.
---
The quoted passage is the coordinate. The note is the instruction. The separator keeps the pairs from bleeding into each other. The model locates each passage in its last output via exact-text match — a thing language models do extremely well — and applies the note in place. It does not regenerate the paragraphs around your quoted passage, because you didn't quote them.
This pattern works in ChatGPT, Claude, Gemini, Cursor's chat mode, and any local model large enough to hold the original answer in context. It is not a trick. It is not a jailbreak. It is the shape of the data the model wanted in the first place.
The full walkthrough — including the manual recipe you can do in a plain-text scratch buffer in ten minutes — is in the guide. If you want to see every method ranked, read the field manual on commenting Markdown in 2026. If the wall you keep hitting is specifically with Cursor on a long README, that has its own playbook.
The cruel twist
ChatGPT will almost certainly ship a native "edit only this passage" surface inside of a year. Claude probably first. Cursor maybe already by the time you read this. The chat box as you know it will get a lightweight comment layer bolted onto the side of the output bubble, and the tweet about it will be "we built this because the chat box isn't great at targeted edits."
Fine. Until then: paired passages, separators, one round-trip. OMGfixMD exists because the manual recipe takes ten minutes and we were tired of doing it.
OMG.