Freelancers, agencies and content teams who draft with AI and need copy that reads like a person wrote it, and non-native English writers whose own writing gets wrongly flagged by AI detectors. Students use it too, and are responsible for following their institution's rules.

Our own detector scores every rewrite, and a good rewrite shows 75 to 100% human there. Outside detectors such as Originality.ai, GPTZero, Copyleaks and Turnitin each use their own models and change them often, so no tool can promise what they will say, and anyone who does is selling you something.

No. There is no documents table in our database. Your text lives in server memory for the length of the request and is discarded when the response finishes. All we store per user is an email, a subscription ID, and a remaining word balance.

Sentence meaning is preserved. What changes is rhythm, sentence length variation, and word choice, since those are the statistical signals detectors read. Academic mode additionally locks terminology and citations so technical accuracy is not disturbed.

That depends on the rules you are working under, and those rules are yours to know and follow. Plenty of legitimate uses exist: agencies polishing AI first drafts, non-native speakers whose natural writing gets falsely flagged, and writers who use a model for structure but want the prose to sound like them. Detectors also produce false positives on genuinely human writing, and that is a real problem people use this to solve.

The counter resets. The Free plan resets on the 1st of each month. Paid plans count rewrites over a rolling 24 hours, so each one frees up a day after you used it. You can upgrade at any point.