Your resubmit got a new AI writing snapshot—stop measuring it with the old attempt badge
Your resubmit got a new AI writing snapshot—stop measuring it with the old attempt badge
In English-medium programmes across India and Southeast Asia, instructors often say revise and resubmit. After the second upload, group chats paste the attempt-1 AI writing badge next to the attempt-2 file and conclude the rewrite “did nothing.” This post is about attempt-snapshot literacy: each submission is usually a new scored snapshot, not a sequel to the old card.
It does not promise any score, and it is not a detection-evasion guide.
Common mismatches
Students glue the attempt-1 portal integer onto the attempt-2 draft; reuse the first downloaded PDF highlight map to judge the second upload; or screenshot “no change” while the queue is still processing. Any of those can invent a false story that the rewrite failed.
Alignment table
Before arguing the delta, match: attempt / submission number, report timestamp, filename and page count, tool name. If the attempt IDs disagree—stop. You may be comparing two different materials.
Check | Attempt 1 | Attempt 2 |
|---|---|---|
Attempt / submission ID | Write it down | Must match the file you are discussing |
Report time | Card + PDF export | New card + new export |
Highlight map | Old PDF only | Download again; do not recycle |
Five steps before you say “it didn’t work”
Record badge numbers, report IDs, and times for both attempts.
Confirm the LMS has finished generating a report for the second upload (processing ≠ frozen old number).
Download the full attempt-2 PDF; do not reuse the attempt-1 export.
Write down which sections the school asked you to revise—movement outside that scope is not proof of misconduct.
In office hours, say the attempt number first, then the numeric difference.
Similarity still answers a different question
A low similarity score does not cancel a new AI writing indicator. A new indicator does not prove plagiarism. Mixing attempts adds noise to both stories.
If your school already required edits: focus revision on the flagged continuous prose in the file that will be rechecked. Keep citations, tables, and verified facts unless they are wrong. For a document-level workflow, see HumanPen. It still does not promise what the next LMS card will show.
Takeaways
Each attempt is usually a new snapshot, not a continuation of the old badge.
Align attempt IDs and timestamps before arguing whether the rewrite “worked.”
No honest tool should be described as a guaranteed pass.
— Charles Li · HumanPen · https://humanpen.net
Join the Conversation
Share Your Thoughts
Related Articles
*% Versus 21%: Do Not Over-Read the First Visible AI Number
The AI writing indicator does not always show a tidy percentage. A display band — not a gavel — explains *% and ~21%.
Authorship Flag vs AI Detection: Two Systems People Mix Up in Reports
Authorship and AI detection answer different questions with different methods. How to read each flag without mixing them up.
The AI Badge Did Not “Miss” Your Rewrite—It May Be Scoring an Older Upload
An AI writing percentage scores a submitted snapshot. Version mismatch — not a stubborn detector — explains most “I fixed it and the score stayed” nights.