Direct answer
You can check a paper for plagiarism and AI writing for free by combining three layers: your university library's licensed similarity tool, a free limited-scan checker for a rough similarity read, and a manual AI-tell review of your own draft. None of these free routes reproduces what your instructor sees, because Turnitin's similarity report and its AI writing indicator are separate signals, and the AI indicator only appears when the institution enables it [1]. Free checks are useful for catching obvious problems early, but they are screening tools, not a preview of your final Turnitin result.
What free ways are there to check a paper for plagiarism and AI writing, and how reliable are they?
The genuinely free options fall into four groups, and each has a different reliability ceiling. Library-licensed tools are the strongest free route, because many universities subscribe to a similarity checker and let enrolled students run a limited number of drafts through it before final submission. Free web checkers typically cap you at a few hundred to a few thousand words per scan, which means a 6,000-word dissertation gets checked in fragments rather than as a whole document.
Free AI checkers are the weakest of the four. Turnitin's own guidance explains that AI writing detection is a probabilistic estimate of how likely a passage is to have been generated by a large language model, and that no detector is fully accurate [2]. That matters because detectors are known to flag formulaic academic prose — a literature review built from standard transition phrases can read as "AI-like" even when a human wrote every word [2].
Manual review is free, unlimited, and underrated. Read your draft aloud, look for sentences that are uniformly medium-length and hedged, and check whether every borrowed idea carries a citation. This will not produce a number, but it reliably surfaces the two things that actually cause trouble: uncited borrowed text and passages you cannot explain in your own words.
What you should not expect from any free tier is a real score. Free tools index different corpora, apply different thresholds, and rarely search the same proprietary student-paper repository your institution uses, so their percentages are directional at best [2].
Why do free plagiarism and AI checkers give different results from Turnitin, and what do they miss?
The biggest gap is the index, not the algorithm. Turnitin's similarity check compares your text against its own repository of previously submitted student papers plus internet and publication sources, and free checkers generally cannot search that student-paper archive at all [3]. If your draft borrows from an essay a student submitted to another university three years ago, a free web checker may return a clean result while your institution's check lights up.
The second gap is configuration. Institutional reports usually exclude bibliographies, quoted material, and matches below a set word count, and those exclusions change the percentage your instructor actually sees [3]. A free checker that counts every three-word overlap will show a scarier number than Turnitin, while one that ignores short matches will show a friendlier one — neither matches your institution's settings.
The third gap is how AI detection is reported. Turnitin shows an asterisk in place of an exact figure when its AI detection confidence falls below its threshold, which signals a low-confidence result rather than a confirmed one [1]. Free AI checkers rarely communicate uncertainty this way, so students often overreact to a precise-looking percentage that the underlying model does not actually support.
Finally, remember that similarity and AI are independent. A paper can have a low similarity score and a high AI indicator, or the reverse, because the two systems measure completely different things [1]. Treating one number as a proxy for the other is the most common misreading students make.
How can you verify your real Turnitin similarity and AI scores before you submit?
Start by reading any report you do get as evidence rather than as a verdict. A Turnitin similarity report highlights each matched string and names the source it matched, so the useful information is which passages matched and whether they were properly cited — not the headline percentage [4]. A 22% score dominated by correctly quoted and referenced material is a very different situation from a 9% score built on unattributed paraphrase.
Then separate the two questions you are actually asking. "Did I borrow text without attribution?" is a similarity question, answered by reading matches in context [4]. "Does this draft read as machine-generated?" is an AI question, answered by the AI writing indicator, which your instructor may or may not have enabled on the assignment [1]. Answering them one at a time stops you from chasing the wrong fix.
If your free checks left you uncertain, the honest move is to get a real preview rather than guess. That means running your draft through the same class of similarity and AI reporting your institution uses, before the deadline rather than after, so you have time to rewrite flagged passages and re-check.
Practical habits that reduce risk at zero cost: keep your drafts and revision history, cite as you write instead of retrofitting citations, and paraphrase by closing the source and writing from your own notes. These do not replace a real check, but they shrink the surface area a real check can flag.
Free checkers tell you something might be wrong; they cannot tell you what your instructor will see. If you want the actual Turnitin AI and similarity reports on your own draft — the same report format your professor opens in the LMS — turnitin0 delivers both PDFs in one check, typically in under 15 minutes, without adding your file to Turnitin's student paper database. Students use it to preview scores, read the flagged passages, and fix them while there is still time.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
FAQ
Is there a completely free way to check for plagiarism and AI writing?
Yes, but only in a limited sense. Your university library's licensed similarity tool is free to you and is the closest free option to what your instructor sees, while public web checkers are free but index different sources and cap your word count [3]. Free AI checkers exist too, but they give probabilistic estimates rather than confirmed results [2].
Why is my free checker's score different from my Turnitin score?
Because the tools search different indexes and apply different exclusions. Free checkers usually cannot search Turnitin's student-paper repository, and institutional reports often exclude bibliographies and quoted material, which shifts the visible percentage [3]. The same draft can legitimately produce several different numbers.
Does a low similarity score mean my paper is safe from an AI flag?
No. Similarity and AI writing detection are separate indicators, and the AI indicator is only shown when the institution enables it on the assignment [1]. A draft can show minimal matching text and still carry a high AI writing percentage.
What does the asterisk next to a Turnitin AI score mean?
Turnitin displays an asterisk instead of an exact percentage when its AI detection confidence sits below its threshold. That marks a low-confidence signal rather than a confirmed finding, so it should be read with caution rather than treated as proof [1].
How should I read a similarity report if I get one?
Read the matches, not just the number. Each highlighted passage is tied to a named source, so the real question is whether that passage is properly quoted and cited [4]. Fixing attribution on flagged passages matters far more than lowering the headline percentage.