Can I use the Turnitin AI detector myself?
No. It is licensed to institutions, the report is generated for staff, and there is no student version. That asymmetry is the entire problem.
The short answer is that Turnitin is sold to institutions rather than to individuals. Access comes through your institution's account, the AI indicator is generated on submission, and the report it appears on is visible to staff.
Some institutions enable a draft or practice submission area where students can see a similarity report. Whether the AI indicator is shown there is a per-institution setting, and in most configurations it is not.
So the realistic position: a score about your writing exists, it was generated automatically, and you cannot see it.
Why that asymmetry matters
Every useful action happens before submission.
After submission you are discussing a number somebody else generated, on a document you can no longer change, without having seen the report. Before submission you can look at what your writing does to a classifier and fix the paragraphs that read flat.
Same information, completely different value depending on when you have it. See what happens when Turnitin flags AI.
What a second detector gets you
Not a prediction of Turnitin's number. Different tool, different threshold, different segmentation, so the figures will not match and anybody promising otherwise is guessing.
What it gets you is the shape: which paragraphs in your document carry the properties every detector reads as machine-written. That part is stable across tools, because they are all reading the same underlying regularity. A section that comes back flat in one comes back flat in the other.
The number to compare tools on
How often a detector is wrong about genuine human writing, since that is the error that arrives attached to somebody's work.
Turnitin publishes 4%. Ours is 0.4%, which is 99.6% specificity, measured across 15,900 documents of real academic writing.
Turnitin also states that documents scoring below 20% show a higher incidence of wrong flags, and their own analysis found 54% of falsely flagged sentences sit directly beside a genuinely flagged one, with 26% two sentences away.
What to run instead, and how
Score the full document rather than an extract, since every detector is more reliable on more text. Read the per-paragraph scores rather than the document figure, because that is what tells you where to work. Then fix the two or three that sit highest: vary the sentence lengths, cut the connective openings, put the specific detail back.
Ours is free and unlimited with no account, takes .pdf, .docx and .tex, has no word cap, scores every paragraph separately, and never stores your text. There is no institution to go through and nothing to ask permission for.