Charles Li / HumanPen
When a tutor mentions a high AI-writing score, the useful first move is not to argue the number. Confirm that the file they scanned is the same file you submitted through the LMS. Different drafts, an old cloud export, or a PDF that still contains the assignment prompt can produce a report that does not match your final essay.
Write down four fingerprints: LMS filename, byte size, upload timestamp, and the exact sentence the tutor quoted. If that sentence is absent from your final DOCX, you are looking at a version mismatch, not a verdict on your thinking. Chat screenshots on WhatsApp or Messenger often omit the tool name and date; treat them as a tip to open the LMS, not as the official record.
Open the institutional email and the assignment page before you draft any defence. Reply only after both of you are looking at the same notice.
Bring dated outlines, reading notes, earlier drafts, and a short note if you used translation or grammar tools. Many university policies already say a detector score alone cannot justify a penalty. Process evidence turns a tense meeting into a concrete review of how the essay was built.
Ask politely for the full highlighted report and the tool name. A bare percentage in email is not enough to discuss. Read highlights like a map: which sections cluster, and do they sit on definitions, literature summary, or conclusion? Careful textbook-style English by multilingual writers is flagged more often in published studies; a high number is a reason to look closer, not proof of misconduct.
After a short confirmation message in chat, move to institutional email or office hours. Point at concrete paragraphs rather than debating the headline score. Agree in writing on next steps and deadlines after the meeting so spoken promises do not drift.
Edit only the continuous highlighted stretches inside your real DOCX, PPTX or TXT. Keep citations and structure. Do not pay for services that promise a guaranteed pass or an undetectable essay — honest tools cannot promise that, and detectors change. Dumping a whole chapter into an anonymous rewriter usually makes the paper trail worse.
https://humanpen.net">HumanPen is for document-level revision of flagged passages after you have read the report as a map. It does not replace process evidence and it does not guarantee the next score.
Send a short confirmation email the same day: which file is treated as authoritative, which stretches will be revised, and by when. Spoken agreements drift quickly across time zones and busy term weeks.
If you drafted first in Hindi, Tamil, Mandarin or another first language and then translated, say so plainly. Translation changes the statistical surface of a text; explaining it is part of an honest process, not an admission of misconduct.
Avoid parallel WhatsApp debates over screenshots while the institutional clarification is open. One channel, one original notice, one file fingerprint.
Document-level revision support: https://humanpen.net">HumanPen (https://humanpen.net).