Direct answer
Yes — text produced by Writer's AI writing platform can be flagged by Turnitin's AI writing detector, because detection targets statistical patterns in the prose rather than the brand of tool that produced it. Turnitin examines submitted text for machine-like predictability and reports the percentage of qualifying prose it classifies as AI-generated, displaying a score only once that share reaches its confidence threshold [1]. For a student who drafted an assignment with Writer's Palmyra models or its writing assistant, the practical question is not whether detection is possible but how much of the draft is likely to be flagged. That is a question you can answer before you submit, not after.
Introduction
Yes — text produced by Writer's AI writing platform can be flagged by Turnitin's AI writing detector, because detection targets statistical patterns in the prose rather than the brand of tool that produced it. Turnitin examines submitted text for machine-like predictability and reports the percentage of qualifying prose it classifies as AI-generated, displaying a score only once that share reaches its confidence threshold [1]. For a student who drafted an assignment with Writer's Palmyra models or its writing assistant, the practical question is not whether detection is possible but how much of the draft is likely to be flagged. That is a question you can answer before you submit, not after.
Does Turnitin detect text written by Writer AI?
Turnitin does not maintain a list of "banned" AI tools, and it does not need one. Its detector evaluates the text itself, so output from Writer is treated the same way as output from any other generative model: the system scores the document and then highlights the specific segments it believes were machine-written [2]. That means a Writer draft is not safe simply because Writer is a business-oriented platform rather than a consumer chatbot. If the prose carries the statistical signature the model associates with AI generation, it will be surfaced in the report.
The score you see is a proportion, not a verdict on authorship. Turnitin's guidance describes the AI writing percentage as the share of qualifying text flagged, which is why a 100% score on a short passage can coexist with substantial human work elsewhere in the document [2]. Students often misread this as an accusation about the whole assignment. It is more accurate to read it as a map of which paragraphs need attention.
There is also a visibility threshold worth knowing. Turnitin only displays a numeric AI score when enough of the document is flagged; below its confidence threshold the report shows an asterisk instead of a precise figure, signalling a low-confidence signal rather than a clean bill of health [1]. A *% result is therefore not proof that your Writer-assisted draft passed scrutiny — it means the detector found too little to report confidently. If your institution reads reports strictly, that distinction matters.
Why does Writer-generated text still get flagged even after I edit it?
Because the features detectors measure survive most of the edits students actually make. AI detection models weigh statistical predictability, uniform sentence rhythm, and low burstiness — the variation in sentence length and structure that human writers produce naturally [3]. Swapping synonyms, reordering clauses, or running a paragraph through a paraphraser leaves those underlying patterns largely intact, so the text keeps reading as machine-generated to the classifier even though it looks different to you.
Paraphrasing tools are a particular trap. Many of them rewrite at the token level while preserving the same predictable word choices and the same flat cadence, which is precisely the signal being measured [3]. A student who paraphrases a Writer paragraph three times may end up with three differently worded passages that all score the same. Effort and score reduction are not the same thing.
What genuinely moves the number is substantive rewriting: varying sentence length deliberately, introducing specific evidence, examples, and citations tied to your own reading, and letting some sentences run long while others stay short [3]. Turnitin's own interpretation guidance notes that human editing reduces flagged segments but does not automatically erase them, and that scores can shift between resubmissions of lightly edited text [2]. The practical takeaway is that edits should change how the argument is built, not just how the words are ordered.
How can I check a Writer draft's Turnitin AI score before submitting?
The only reliable way to know your score is to see the report itself before the deadline, and that requires a check that mirrors what your institution runs. Turnitin's guidance for instructors frames the AI writing report as a starting point for conversation rather than a final judgement, and notes that students benefit from seeing which segments are flagged while they can still revise [4]. Waiting for the institutional report removes that option entirely — by then the score is part of your record.
Institutional rules on pre-submission checking vary, so read your own handbook first [4]. Some departments explicitly permit draft checks through external tools; others prohibit uploading coursework anywhere outside the official submission portal. Knowing which category you fall into is part of using any checker responsibly, and it is the difference between a sensible precaution and a policy breach.
When pre-checking is permitted, the useful output is not a single number but a segment-level view. A report that shows which paragraphs triggered the detector, alongside a similarity summary, tells you exactly where to rewrite and whether you also have a citation problem [4]. That combination — AI flags plus matched sources — is what lets you fix a draft in one revision pass instead of guessing which paragraph to attack.
If you would rather see the flags while there is still time to act, turnitin0 gives you that segment-level view before you submit. Upload your .docx, .pdf, or .txt file and you receive two PDFs in one checkout — a Turnitin AI detection report and a similarity report, matching what instructors see in their LMS. Files are checked without being added to Turnitin's student paper database, turnaround averages under 15 minutes, and 100,000+ reports have been delivered to 20,000+ students across the US, UK, Canada, Australia, New Zealand, and Ireland. A single check is $3.80, with a 10-check pack at $2.75 each and no subscription.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Is Writer AI detectable by Turnitin?
Yes. Turnitin scores the text rather than the tool, so Writer output is evaluated on the same statistical features as any other generative model's prose [1]. A Writer draft with predictable rhythm and uniform sentence structure is likely to produce flagged segments.
What does a *% score on my report mean?
It means Turnitin detected AI signals but not enough of them to report a confident percentage, since the numeric score appears only once flagged text crosses its threshold [1]. Treat it as a low-confidence signal, not a pass.
Will rewriting in my own words remove the flags?
Only if the rewriting changes the underlying structure. Varying sentence length, adding specific evidence, and rebuilding the argument lower the score; synonym swaps and paraphrasing tools generally do not [3].
Can I check my draft before submitting it?
That depends on your institution's policy, which Turnitin notes varies between departments [4]. Where draft checks are permitted, a segment-level report shows you which paragraphs need revision while you can still revise them.
Does using Writer count as academic misconduct?
Turnitin's report is an indicator, not a finding of misconduct, and instructors are advised to interpret it in context [2]. Whether a given use of AI writing tools is allowed is set by your course policy, not by the detector.