QuillBot vs Turnitin AI detector: a free consumer check against an institutional licence, on the same 14 fields
QuillBot's AI Detector is free for 6 scans a day of up to 1,200 words with no sign-up, and Turnitin's AI writing detection "is only available to customers that license Turnitin Originality", so the pair is the one a student meets twice: once on their own screen and once inside the Similarity Report their instructor sees (quillbot.com/ai-content-detector, guides.turnitin.com, read 8 November 2026). The evidence does not overlap: Turnitin carries 5 records, from 41 of 54 documents correct and 0 false accusations in 18 in Weber-Wulff 2023 to all 40 fully AI-generated papers scored under 20% in Van Vlasselaer 2026, and QuillBot carries 2, Kar 2024 and Alshammari and Rao 2025, both printing displayed scores rather than rates, with no study that ran both (DOI 10.1007/s40979-023-00146-z, Tables 5 and 9; DOI 10.1007/s40979-026-00226-w, Table 4; DOI 10.1177/02537176241247934, Table 1; arXiv 2507.17944, Table III; read 8 and 9 November 2026). The two cards are read side by side below on the same 14 fields, with the vendor's wording and the page it came from.
Updated Fields checked 8 November 2026
Short answer
QuillBot's detector is free for 6 scans a day of up to 1,200 words with an 80-word minimum and no sign-up under that length, states 6 languages and six browser and app surfaces, publishes no API and no retention statement, and removes the word limit on Premium at $8.33 a month billed annually; Turnitin's AI writing detection has no free tier and no public price, takes 300 to 30,000 words of prose in .docx, .pdf, .txt or .rtf, states 7 languages, is delivered inside the Similarity Report through the LMS, and keeps the AI result as part of that report (quillbot.com/ai-content-detector, quillbot.com/premium, guides.turnitin.com, read 8 November 2026). Turnitin has four measured false positive counts, all zero (0 of 18, 0 of 42, 0 of 10 and 0 of 40), and the largest measured miss in the index, all 40 Deep Research papers in Van Vlasselaer 2026; QuillBot has no measured rate on human text and a vendor-submitted RAID leaderboard entry of 30 January 2025 at 99.62% (Weber-Wulff 2023; Walters 2023; Perkins 2024; Van Vlasselaer 2026; github.com/liamdugan/raid, read 9 November 2026). Neither vendor publishes an accuracy figure for its own tool that a study has paired. No detector score should be the only evidence in a misconduct, hiring or publishing decision.
How do QuillBot and Turnitin compare on the 14 fields?
| Field | QuillBot | Turnitin |
|---|---|---|
| 1. Name, vendor, official domain | AI Detector by QuillBot, QuillBot, quillbot.com | Turnitin AI writing detection, Turnitin, LLC, turnitin.com |
| 2. Modalities detected | text, image | text |
| 3. Generators covered | GPT-6, GPT-5, ChatGPT, Gemini, Claude, Llama and "other LLMs" | 33 model versions from Amazon Nova 2 Lite and Claude-3-Haiku to GPT-5.4-pro, Grok-4.1, Llama-4-Maverick and Mistral-large-3; Spanish model GPT-3.5 and GPT-4; Japanese model GPT-4, GPT-4o and GPT-4o-mini |
| 4. Languages supported | 6: English, French, Spanish, German, Dutch, Portuguese | 7: English, Spanish, Japanese, Modern Arabic, Dutch, German, Turkish |
| 5. Input limits | Minimum 80 words per scan; Free up to 1,200 words per scan; Premium no word limit; paste or upload a document (file types and size not published); Premium bulk upload | At least 300 words of prose and no more than 30,000 words; file under 100 MB; .docx, .pdf, .txt, .rtf; lists, headers, references, tables, formulas and code excluded from scoring; unsupported-language submissions not processed |
| 6. Free tier | Up to 1,200 words per scan, up to 6 scans per day, 1 free explainer card; no sign-up for scans under 1,200 words | None: "only available to customers that license Turnitin Originality (or the AI writing add-on for iThenticate 2.0)" |
| 7. Paid plans | Premium $8.33 a month billed annually, bundled with the writing suite: no word limit, unlimited scans and explainer cards; Team Plan, price not shown | No public plan: licensed with Turnitin Originality or as the iThenticate 2.0 add-on through an account manager; no price published |
| 8. API | ||
| 9. Integrations | Google Chrome extension, Microsoft Edge extension, macOS app, Word add-in, Android keyboard and app, iOS keyboard and app | LMS (Moodle, Blackboard, Canvas and others) through the Similarity Report, with no AI indicator in the LMS interface; Microsoft Teams for Feedback Studio only |
| 10. Vendor accuracy and false positive claims | "Detect AI text reliably with a 99% detection rate, according to independent evaluations from RAID"; "our model tends to classify texts as human written, which reduces false positives", no numeric rate | No accuracy figure; false positive rate "under 1% for documents with over 20% of AI writing", tested on "over 700,000 additional academic papers that were written before the release of ChatGPT"; ELL writers 0.014 and native writers 0.013 in the vendor's research |
| 11. Independent studies | 2 records: Kar 2024, Alshammari and Rao 2025 | 5 records: Weber-Wulff 2023, Walters 2023, Perkins 2024, Van Vlasselaer 2026, and the vendor claims record |
| 12. Data retention and privacy statement | No retention statement on the detector page or in the pages read | No separate repository for AI detection; the AI percentage and highlighted segments "are retained as part of the similarity report"; full deletion of a submission on request, no partial deletion of the AI component |
| 13. Launch year and ownership | Not stated; part of the Learneo family of businesses, stated on scribbr.com, a sister Learneo brand | 4 April 2023; Turnitin, LLC, no parent stated on the pages read |
| 14. Rankings the card appears in | Independently tested, languages, studies by content type, platform integrations, minimum input, free checks per day, generators covered; unlimited table of input limits | Independently tested, false positive rate, free tiers, languages, input limits, adversarial robustness, false negative rate, platform integrations, file types, minimum input, LMS coverage, generators covered, non-English evidence |
Which studies tested QuillBot or Turnitin?
None tested both. Weber-Wulff 2023 ran 14 tools on 54 documents between March and May 2023, Turnitin among them and QuillBot not, and the authors note that Turnitin knew the documents were part of a test suite (DOI 10.1007/s40979-023-00146-z, Tables 5, 9 and 10, read 8 November 2026). Walters 2023 submitted 126 documents to 16 detectors, Turnitin among them (De Gruyter 10.1515/opis-2022-0158, Table 4, read 8 November 2026), and notes that the student papers could have been used to train the Turnitin detector. Perkins 2024 ran seven detectors on 104 AI samples before and after six adversarial techniques (DOI 10.1186/s41239-024-00487-w, Tables 7, 8 and 10, read 9 November 2026), and Van Vlasselaer 2026 ran four detectors on 160 papers in four classes (DOI 10.1007/s40979-026-00226-w, Tables 4 and 5, read 9 November 2026); the Turnitin claims record collects the vendor's own statements from 2023 to 2026 and is not an independent study. QuillBot's two records are Kar 2024, one ChatGPT 3.5 article and three paraphrases through ten free detectors with no human text (DOI 10.1177/02537176241247934, Table 1, read 9 November 2026), and Alshammari and Rao 2025, six detectors on 49 DeepSeek-V3 answers, 49 human answers and paraphrased sets, printed as mean displayed scores (arXiv 2507.17944, Tables III to V, read 9 November 2026). QuillBot's RAID leaderboard entry is a vendor submission read from the repository and is listed as an entry, not a record (leaderboard question).
| Source | QuillBot | Turnitin |
|---|---|---|
| Weber-Wulff 2023, 54 documents (Tables 5, 9 and 10) | Not tested | 41 of 54 correct (76%), rank 1 of 14; 0 false accusations in 18 human documents; 10 of 36 AI documents undetected (27.8%), all among manually edited and QuillBot-paraphrased texts |
| Walters 2023, 126 documents (Table 4) | Not tested | 100% correct: 0 of 42 human papers flagged, 100% on 42 GPT-3.5 and 42 GPT-4 papers |
| Perkins 2024, 104 AI and 10 human samples (Tables 7, 8 and 10) | Not tested | 61% baseline mean accuracy, rank 2 of 7; 50% on unmanipulated AI text falling to 7.9% after adversarial techniques, the largest drop; 0 false accusations in 10 human samples; 88 of 104 AI samples undetected |
| Van Vlasselaer 2026, 40 papers per class (Tables 4 and 5) | Not tested | All 40 fully AI-generated papers scored under 20%; 60.0% accuracy on hybrid and 50.0% on humanised papers; all 40 human papers classified correctly |
| Turnitin claims record, vendor statements 2023 to 2026 | Not applicable | Document false positive rate under 1% for documents with over 20% AI writing, validated on 800,000 (May 2023) and over 700,000 (current FAQ) pre-ChatGPT papers; sentence-level rate about 4% (2023); scores of 1% to 19% shown as an asterisk; minimum raised from 150 to 300 words; no recall or accuracy figure |
| Kar 2024, one ChatGPT 3.5 article and three paraphrases (Table 1) | 100% AI on the original; 90 AI after QuillBot's own paraphrase; 100 after Grammarly and ChatGPT paraphrases | Not tested |
| Alshammari and Rao 2025, 49 answers per set (Tables III and V) | Mean AI score 98.4% on DeepSeek-V3 answers; mean human score 100% on 49 human answers; 95.4% on DeepThink text, 58% after a humanized paraphrase | Not tested |
| RAID leaderboard entry, vendor-submitted 30 January 2025 (read 9 November 2026) | 99.62% at a 5% false positive rate across the RAID generations, 2,497 missed, AUROC 0.9962 | No entry |
| Vendor statement, quillbot.com (read 8 November 2026) | "99% detection rate, according to independent evaluations from RAID"; no false positive rate | |
| Vendor statement, guides.turnitin.com (read 8 November 2026) | "under 1%" false positive rate for documents with over 20% AI writing; ELL writers 0.014, native writers 0.013 |
Results as each study reports them: DOI 10.1007/s40979-023-00146-z, De Gruyter 10.1515/opis-2022-0158, DOI 10.1186/s41239-024-00487-w, DOI 10.1007/s40979-026-00226-w, DOI 10.1177/02537176241247934, arXiv 2507.17944, github.com/liamdugan/raid, read 8 and 9 November 2026. "Not tested" means the paper did not run that tool. The false positive rate ranking sorts Turnitin on the Walters 2023 zero and lists QuillBot among the cards with no measured rate; the false negative rate ranking carries Turnitin's 0 of 84 from the same study beside the Van Vlasselaer miss in the block.
What do the two cost over 12 months?
| Plan | Monthly as stated | Annual as stated | Inclusions as stated | Checked on |
|---|---|---|---|---|
| QuillBot free detector | $0 | Up to 1,200 words per scan, 6 scans a day, 1 free explainer card; no sign-up under 1,200 words | 8 November 2026 | |
| QuillBot Premium | $8.33 per month, billed annually (monthly price not shown with the annual toggle on) | $99.96 at the stated monthly rate times twelve | Full access to the AI Detector with no word limit, unlimited scans and explainer cards, bundled with the paraphraser, grammar checker and the rest of the suite; no detector-only plan | 8 November 2026 |
| QuillBot Team Plan | Price not shown | Price not shown | "Explore Team Plan" on the premium page | 8 November 2026 |
| Turnitin AI writing detection | No price published | No price published | Included for customers that license Turnitin Originality; Similarity, Feedback Studio and Originality Check customers are told to speak to their account manager | 8 November 2026 |
| iThenticate 2.0 AI writing add-on | No price published | No price published | "the AI writing add-on for iThenticate 2.0", for the publishing and research product, by account manager | 8 November 2026 |
Prices in USD as shown on quillbot.com/premium and the Turnitin capabilities FAQ, checked 8 November 2026. QuillBot's $99.96 is the index's multiplication of the stated $8.33 by twelve; Turnitin publishes no figure for an individual, an instructor or an institution on any page read, so a 12-month cost for the detector cannot be stated here and sits with the institution's licence.
One of the comparisons in the compare hub. The cards are QuillBot and Turnitin, and the home board answers which AI detectors exist and who has tested them? A student who ran a draft through QuillBot before submission and wants to know what the same draft will meet in Turnitin, and an instructor who wants to know what the free tool the student used can and cannot show, are the readers this pair is written for; the institution-only view explains how Turnitin is bought and the free checks per day ranking places QuillBot's 6 scans beside the other daily allowances.
Which card fits which reader?
- Choose QuillBot ifyou want to check a draft yourself today: free for 6 scans a day of up to 1,200 words with no sign-up, where Turnitin's detector is visible only to instructors and administrators at a licensing institution (quillbot.com/ai-content-detector, guides.turnitin.com, read 8 November 2026).
- Choose Turnitin ifa measured false positive count matters: four zeros on human text across Weber-Wulff 2023, Walters 2023, Perkins 2024 and Van Vlasselaer 2026, where QuillBot's two records print no rate on human text (Turnitin card; QuillBot card).
- Choose QuillBot ifthe result needs to distinguish "Generated and refined by AI" from "Written by a human and refined by AI": Turnitin returns a percentage with scores of 1% to 19% shown as an asterisk (quillbot.com/blog, Turnitin claims record).
- Choose Turnitin ifyour text is in Japanese, Modern Arabic or Turkish: QuillBot states six languages and Turnitin seven, with Spanish and Japanese on separate models (quillbot.com/blog, guides.turnitin.com, read 8 November 2026).
- Choose QuillBot ifyour text is under 300 words: Turnitin requires at least 300 words of prose and QuillBot 80 (guides.turnitin.com, quillbot.com/ai-content-detector, read 8 November 2026).
- Choose Turnitin ifyou need a stated retention term for the result: the AI percentage is kept inside the similarity report with a deletion route on request, where QuillBot's detector page states nothing (guides.turnitin.com, read 8 November 2026).
- Choose QuillBot ifyou want the check inside Chrome, Edge, Word, macOS or a phone keyboard: six surfaces against an LMS report with no AI indicator in the LMS interface (quillbot.com/premium, read 8 November 2026).
- Choose Turnitin ifyou want to know what ChatGPT Deep Research output does to a detector, because the result is on record: all 40 such papers scored under 20% in Van Vlasselaer 2026, where QuillBot has not been tested on it (DOI 10.1007/s40979-026-00226-w, Table 4, read 9 November 2026).
How does the evidence compare, tool by tool?
QuillBot
- Register ID
- AIDI-D-0007
- Study records
- 2 in the index: Kar 2024, Alshammari and Rao 2025
- Shared with Turnitin
- 0
- Vendor accuracy claim
- Detect AI text reliably with a 99% detection rate, according to independent evaluations from RAID.
- Free tier
- Up to 1,200 words per scan, up to 6 scans per day, 1 free explainer card; no sign-up for scans under 1,200 words
- Languages
- 6 stated
Turnitin
- Register ID
- AIDI-D-0002
- Study records
- 5 in the index: Weber-Wulff 2023, Walters 2023, Perkins 2024, Van Vlasselaer 2026, Turnitin claims
- Shared with QuillBot
- 0
- Vendor accuracy claim
- None published
- Free tier
- None. AI writing detection is only available to customers that license Turnitin Originality (or the AI writing add-on for iThenticate 2.0).
- Languages
- 7 stated
How this comparison is built
- Fields
- The 14 fields of every card, in card order
- Shared study records
- 0
- Fields checked
- 8 November 2026
- Records in the studies index
- 35 records
Both columns are drawn from the same 14 fields in the index data file, read on 8 November 2026 from the vendor page recorded on each card, and from the studies index, where each result is copied as the paper reports the figure, with the base stated. No record ran both tools, so the evidence table lists each record with "Not tested" in the other column, the Turnitin claims record in a row labelled as vendor statements, QuillBot's RAID entry in a row labelled as an entry rather than a study, and each vendor's own statement in a labelled row. Turnitin's plans field is empty because the vendor publishes no price, and the cost table says so rather than estimating one. The rules, the recheck schedule (next pass 8 December 2026) and the correction route are on the methodology page; the claim vs evidence gap ranking lists both cards among those with no accuracy claim a study can pair.
Where each figure comes from
Values as each vendor states them on quillbot.com/ai-content-detector, quillbot.com/premium, the QuillBot detector blog post, scribbr.com/about-us, the Turnitin capabilities FAQ and the Turnitin model page, all read 8 November 2026. An empty cell means the card has no published value for the field: neither vendor documents an API for the detector, and QuillBot publishes no retention statement for it. Turnitin's free tier cell is a stated "none", which the free tiers ranking sorts at 0.
Questions about QuillBot and Turnitin
Will Turnitin flag text that QuillBot called human?
No study has run the same text through both, so the index cannot say. Weber-Wulff 2023 found that Turnitin's 10 undetected AI documents were "all among manually edited and Quillbot-paraphrased texts", a result about QuillBot's paraphraser, not its detector (DOI 10.1007/s40979-023-00146-z, Table 10, read 8 November 2026).
Which has a measured false positive rate?
Turnitin: 0 of 18 human documents in Weber-Wulff 2023, 0 of 42 in Walters 2023, 0 of 10 in Perkins 2024 and 0 of 40 in Van Vlasselaer 2026 (Turnitin card, read 10 November 2026). QuillBot's two records print displayed scores, and its page states no numeric rate.
Can a student use Turnitin's detector on their own draft?
No. The capabilities FAQ says the feature is "only available to customers that license Turnitin Originality" and that instructors and administrators see the indicator (guides.turnitin.com, read 8 November 2026); the institution-only view lists the other cards sold the same way.
What does QuillBot's 99% claim rest on?
A vendor-submitted RAID leaderboard entry of 30 January 2025 at 99.62% with 2,497 misses and AUROC 0.9962 (github.com/liamdugan/raid, read 9 November 2026); the RAID 2024 paper itself did not test QuillBot, and the entry is not a study record.
What does Turnitin's "under 1%" rest on?
The vendor's own tests on over 700,000 pre-ChatGPT academic papers, for documents with over 20% AI writing, with a sentence-level rate of about 4% stated in 2023 (Turnitin claims record); no third-party audit of the figure is published, and the measured zeros in four studies are on 18, 42, 10 and 40 human documents.
How much does each cost for a year?
QuillBot Premium is $8.33 a month billed annually, $99.96 for twelve months at the stated rate, and the free tier costs nothing (quillbot.com/premium, read 8 November 2026); Turnitin publishes no price for AI writing detection, which is included in a Turnitin Originality licence sold through an account manager.
Which supports more languages?
Turnitin names 7 and QuillBot 6; Turnitin's paraphraser and bypasser detection is English only (guides.turnitin.com, quillbot.com/blog, read 8 November 2026).
Sources and revisions
Sources
- QuillBot AI Detectorquillbot.com/ai-content-detector
- QuillBot Premiumquillbot.com/premium
- QuillBot detector blog post (languages, result bands)quillbot.com/blog/quillbot-tools/ai-detector
- Scribbr about (Learneo)scribbr.com/about-us
- Turnitin AI writing detection capabilities FAQguides.turnitin.com/hc/en-us/articles/284...
- Turnitin AI writing detection model pageguides.turnitin.com/hc/en-us/articles/282...
- Weber-Wulff 2023doi.org/10.1007/s40979-023-00146-z
- Walters 2023degruyterbrill.com/document/doi/10.1515/o...
- Perkins 2024 (read 9 November 2026)doi.org/10.1186/s41239-024-00487-w
- Van Vlasselaer 2026 (read 9 November 2026)doi.org/10.1007/s40979-026-00226-w
- Kar, Bansal, Modi and Singh 2024 (read 9 November 2026)doi.org/10.1177/02537176241247934
- Alshammari and Rao 2025 (read 9 November 2026)arxiv.org/abs/2507.17944
- RAID leaderboard submissions (read 9 November 2026)github.com/liamdugan/raid/tree/main/leade...
