At UCLA, an AI-detection accusation followed a documented pattern: a Turnitin AI indicator score triggered an integrity referral, and the student cleared the case by producing handwritten planning notes and browser history that documented their research and drafting. The detector output did not change. The process evidence did.
The pattern at UCLA
UCLA handles academic integrity allegations through the Office of Student Conduct, working from the UCLA Student Conduct Code and system-wide policy under UC PACAOS-100. Instructors report suspected misconduct; the office then investigates, offers the student a chance to respond, and either resolves the case administratively or refers it to a hearing. A Turnitin AI indicator score is a starting point for that inquiry, not a finding.
In cases that resolve in the student's favor, the sequence is consistent. A paper is flagged, often with a high AI-writing percentage. The student is contacted. They produce evidence of process: notebooks, drafts, browser history, library records. When that evidence lines up with the finished paper, administrators close the case without a formal finding. UCLA's own faculty resources, along with 2023 UC teaching-center guidance, caution against treating detector output as reliable evidence on its own.
Why handwritten notes and browser history are decisive
A detector produces a probability, not a witness statement. Handwritten notes and browser history are different in kind: they are timestamped, external, and hard to fabricate after the fact. Notes taken during class, outlines sketched before drafting, and search records from the days you were researching all reconstruct a writing process. Reconstruction is what administrators need in order to close a case; a percentage score cannot provide it.
Two features of this evidence do the work:
- Temporal alignment. Browser history and file metadata place your research in the days before submission. If the timeline of visits, searches, and drafts matches the arc of the finished paper, the evidence tells a coherent story.
- Content alignment. Handwritten pages that contain the same theses, quotations, and structural choices as the finished essay show the ideas originated with you, in your own hand, before any final text existed.
Why the detector flagged the paper in the first place
AI detectors measure statistical patterns like perplexity (how predictable word choices are) and burstiness (variation in sentence length). Careful academic writing tends to have lower perplexity because students choose precise vocabulary and standard structures. Peer-reviewed research, including the 2023 Weber-Wulff evaluation of fourteen detectors in the International Journal of Educational Integrity and the Liang et al. Stanford study in Patterns, has documented meaningful false positive rates on human-written text, with sharper effects on non-native English writers.
The score does not identify AI use. It identifies text whose statistical fingerprint resembles the training data the detector was trained on. That is why process evidence, not counter-detection, is the response administrators actually respond to.
What UCLA policy actually requires
Under UCLA's Student Conduct Code and the UC system-wide standard, a finding of academic misconduct requires that the alleged violation be established by a preponderance of the evidence. That is a lower bar than criminal proof, but it is still an evidentiary standard that must be met with evidence, not with a single automated score. UCLA students are entitled to notice of the specific allegation, an opportunity to review the evidence being used against them, and an opportunity to respond in writing or in a meeting.
Ask, in writing: which detector produced the score, what the score was, what threshold the office considers significant, and whether any human review of the flagged text preceded the referral. Answers to those questions define what you need to rebut.
Evidence that tends to close these cases
- Handwritten notes, outlines, and marginalia dated before submission
- Browser history covering the research window, exported before it rolls off
- Google Docs or Word version history showing drafting in real time
- Library database access logs and downloaded PDFs with timestamps
- Email or message threads discussing the paper with peers or a TA
- Prior graded work in the same course that shows your consistent style
If this is you at UCLA
Do not respond to the initial email off the top of your head. Read the allegation carefully, note the deadline, and ask in writing for the specific evidence being used, including the detector name and score. Then gather your process evidence in one folder, ordered by date. When you respond, present the timeline first and address the detector second. Your procedural rights FAQ covers what you are entitled to request before any meeting.
For a similar case at another UC campus that turned on the same kind of evidence, see the UC Davis handwritten draft case. If you are preparing a written response, NotBot's defense package generates a personalized letter, an evidence guide keyed to your writing process, and a hearing brief that addresses the detector that flagged you.
If the proposed sanction is suspension or dismissal, or if you are an international student whose visa status depends on enrollment, consult an education law attorney before your meeting.
Build your UCLA defense package
A personalized response that documents your writing process and addresses the detector that flagged your work.
Get your defense package$49 one-time · Generated in 60 seconds