I’d like to suggest a feature that I believe would significantly improve the editing experience with ChatGPT: a “Strict Edit Mode” that ensures only user-specified sections of a document are changed during revisions.
Currently, when I ask ChatGPT to revise a particular sentence or paragraph, the model sometimes also changes unrelated sections — even those I did not ask to be modified. This behaviour often occurs without warning and can result in the unintended removal or rewording of strong, carefully crafted points. It’s not always obvious that this has happened, which increases the risk of losing important material during the editing process.
As a user, it’s entirely reasonable to assume that if I haven’t requested a change to a paragraph, then it should remain untouched. There’s no intuitive reason to expect that editing one part would silently alter another.
A Strict Edit Mode would allow users to:
Confine edits only to explicitly selected or mentioned text
Prevent automatic rewording of other content for the sake of tone or flow
Reduce the need to manually double-check for unintended changes after each revision
This would be especially helpful for letters, legal documents, personal statements, or anything requiring high precision.
Thank you for your ongoing work — this feature would bring even more trust and control to an already powerful tool.