Reviewed 19 November 20268 detectors5 public criteria

The best AI content detectors, ranked

Copyleaks and GPTZero tie for first on public criteria. No score is proof: test any detector on your own text.

Rank · detector · best forScore /12
  1. 01CopyleaksSchools12
  2. 02GPTZeroTeachers12
  3. 03Originality.aiSEO publishers11
  4. 04PangramEditors11
  5. 05SaplingWriters10
  6. 06ZeroGPTQuick checks9
  7. 07Winston AIScans & OCR9
  8. 08TurnitinUniversities8

The criteria, stated first

Full method
  1. 1Published evidence×2Does the vendor publish how it tests accuracy, including its false-positive rate?
  2. 2Explains its scoreDoes it show which passages drove the score, or only a single number?
  3. 3Free to testCan you run your own test before you pay?
  4. 4Public priceIs the price on a public page, or behind a sales call?
  5. 5Fits a workflowCan it run where the work is done: an API, an LMS, a docs plugin, bulk scans?

The ranking

Who each detector suits, and the verdict

  1. 1

    Copyleaks

    Schools

    Full marks on all five criteria: a published test method, free checks and 30+ languages.

    Best for
    Schools and teams that check many languages or work inside an LMS
    Strengths
    • Free checks up to 25,000 characters without logging in
    • AI detection in 30+ languages
    Limits
    • API and LMS access only on Education and Enterprise plans, by quote
    • Sentence-level insights need a free account
    Cost
    Personal $16.99/month ($13.99 billed yearly). Pro $99.99/month ($74.99 billed yearly). Education and Enterprise by quote. Check price
    12/12
    Evidence 2
    Explains 2
    Free test 2
    Price 2
    Workflow 2
  2. 2

    GPTZero

    Teachers

    The teacher's pick: a free monthly allowance, sentence highlights and published benchmarks.

    Best for
    Teachers, students and anyone who wants a free monthly allowance
    Strengths
    • Renewing free plan, enough to run the full test on this site
    • Highlights the sentences behind a score
    Limits
    • Like every detector, it can flag human writing
    • The free plan includes one advanced scan; the rest are basic
    Cost
    Free: 10,000 words a month. Premium $12.99/month, Professional $24.99/month (billed yearly). Check price
    12/12
    Evidence 2
    Explains 2
    Free test 2
    Price 2
    Workflow 2
  3. 3

    Originality.ai

    SEO publishers

    The pick for content teams: bulk scans, plagiarism and fact checks in one account.

    Best for
    SEO publishers and agencies checking content at volume
    Strengths
    • Bulk scan, shareable reports and team seats
    • Plagiarism, readability and fact checks in the same scan
    Limits
    • Free credits are one-time, not monthly
    • API access only on the Enterprise plan
    Cost
    Free: 10,000 words, one time. Pro $14.95/month ($12.95 billed yearly). Enterprise $179/month. Check price
    11/12
    Evidence 2
    Explains 2
    Free test 1
    Price 2
    Workflow 2
  4. 4

    Pangram

    Editors

    The evidence-first pick: a public model card behind the model, priced for one person.

    Best for
    Editors and researchers who want the method behind the model
    Strengths
    • Publishes a model card and research on how its detector works
    • Checks text, images and video
    Limits
    • No renewing free tier on the pricing page, only a trial
    • The trial starts a paid plan unless you cancel
    Cost
    Individual $20/month after a 7-day free trial. Professional $65/month. API $0.05 per 100 words. Check price
    11/12
    Evidence 2
    Explains 2
    Free test 1
    Price 2
    Workflow 2
  5. 5

    Sapling

    Writers

    The best free quick check for short text, with a pay-as-you-go API behind it.

    Best for
    Writers checking short pieces, and developers who want a metered API
    Strengths
    • Free checks without an account
    • Scores each sentence as well as the whole text
    Limits
    • Says false positives occur, but publishes no rate
    • Free checks cap at 5,000 characters, about 800 words
    Cost
    Free: checks up to 5,000 characters each. Pro $25/month ($12 billed yearly). API metered, $5 minimum. Check price
    10/12
    Evidence 1
    Explains 2
    Free test 2
    Price 2
    Workflow 2
  6. 6

    ZeroGPT

    Quick checks

    Fine for a free first look; too little published evidence to act on alone.

    Best for
    A quick free look, never a decision
    Strengths
    • Free tier
    • Highlights sentences it reads as AI
    Limits
    • Claims high accuracy without a published false-positive rate
    • No test method you can rerun against its claims
    Cost
    Free tier, $0. Personal plans from $14.99/month. API $0.042 per 1,000 words. Check price
    9/12
    Evidence 1
    Explains 2
    Free test 2
    Price 2
    Workflow 1
  7. 7

    Winston AI

    Scans & OCR

    The pick when your text arrives as photos, scans or PDFs that need a report.

    Best for
    Teachers and publishers who check scanned or handwritten pages
    Strengths
    • Reads pictures and handwriting (OCR)
    • Shareable PDF reports
    Limits
    • Free access is a trial, not a tier
    • Accuracy is a headline claim, not a published method
    Cost
    Free 14-day trial (2,000 credits). Essential listed at $18/month billed monthly; promotions change it. Check price
    9/12
    Evidence 1
    Explains 2
    Free test 1
    Price 2
    Workflow 2
  8. 8

    Turnitin

    Universities

    Right only if your institution already pays for it; you cannot test it on your own.

    Best for
    Universities and schools that already grade inside Turnitin
    Strengths
    • Sits inside the LMS and grading workflow teachers already use
    • Publishes guidance on its false-positive rate and how to read scores
    Limits
    • Individuals cannot buy it or run their own test
    • Some universities have switched its AI indicator off
    Cost
    Sold to institutions. No price on its public pages: ask your institution or Turnitin. Check price
    8/12
    Evidence 2
    Explains 2
    Free test 0
    Price 0
    Workflow 2

Flagged your own draft?

Fix the writing with Lettershred

Rewrites your own AI-assisted drafts so they read as useful and human. Assessed on the same five criteria, with its limits.

See Lettershred
01SplitSplits every sentence, holding headings, lists and tables fixed
02RewriteRewrites across several model providers
03CheckChecks each rewrite against the original meaning
Same criteria6/12
Evidence 1
Explains 0
Free test 0
Price 2
Workflow 2
  • Not a detector: it never tells you whether a text is AI
  • For your own content, not for work you are required to write yourself
  • Facts still need an editor's read

Part of Letter's plans, from $100 a month on letter.dev/pricing. Check price

Side by side

Comparison table

#DetectorEvidence×2ExplainsFree testPriceWorkflowTotalCost
01CopyleaksSchoolsPublished testing method with false-positive ratesHighlights AI-written phrases, sentence by sentenceFree checks, 25,000 characters without loginPersonal and Pro priced on copyleaks.com/pricingDocs add-on, extension, MCP; API and LMS by quote12/12Personal $16.99/month ($13.99 billed yearly). Pro $99.99/month ($74.99 billed yearly). Education and Enterprise by quote. Check
02GPTZeroTeachersPublishes benchmarks with its false-positive rateSentence-level highlightingFree plan, 10,000 words a monthPlans priced on gptzero.me/pricingAPI, Chrome extension, team plans, LMS12/12Free: 10,000 words a month. Premium $12.99/month, Professional $24.99/month (billed yearly). Check
03Originality.aiSEO publishersPublishes accuracy data and false-positive ratesSentence-level precision on every plan10,000 words, one timeAll plans priced on originality.ai/pricingBulk scan, team seats, API on Enterprise11/12Free: 10,000 words, one time. Pro $14.95/month ($12.95 billed yearly). Enterprise $179/month. Check
04PangramEditorsModel card and stated false-positive rateHighlights the sections it reads as AI7-day free trialPlans and API rates on pangram.com/pricingAPI, MCP, Docs and Gmail, team plans11/12Individual $20/month after a 7-day free trial. Professional $65/month. API $0.05 per 100 words. Check
05SaplingWritersStates false positives occur; no published rateScores each sentenceFree, 5,000 characters per checkPlans priced on sapling.ai/pricingMetered API, browser extension, team plan10/12Free: checks up to 5,000 characters each. Pro $25/month ($12 billed yearly). API metered, $5 minimum. Check
06ZeroGPTQuick checksAccuracy claims; no published false-positive rateSentence highlightingFree tier, $0Plans priced on zerogpt.com/pricingPay-as-you-go API9/12Free tier, $0. Personal plans from $14.99/month. API $0.042 per 1,000 words. Check
07Winston AIScans & OCRHeadline accuracy claimsPrediction map, sentence by sentence14-day trial, 2,000 creditsPlans on gowinston.ai/pricingAPI, team members, PDF reports9/12Free 14-day trial (2,000 credits). Essential listed at $18/month billed monthly; promotions change it. Check
08TurnitinUniversitiesPublishes its false-positive rate and how it holds itHighlights text in the reportNo individual accessNo public priceMoodle, Blackboard, Canvas and other LMSs8/12Sold to institutions. No price on its public pages: ask your institution or Turnitin. Check

● = 1 point, two dots is the most a criterion gives. Evidence counts double. Prices as listed on 19 November 2026.

Where detectors fail

A score is evidence, not proof

When detectors get it wrong
61.3%
of essays by non-native English writers were flagged as AI, on average across seven detectors. Liang et al., Patterns, 2023
9%
of human text was labelled AI by OpenAI's own classifier, retired in July 2023. OpenAI

Choosing

Questions people ask when choosing

All questions
Which AI content detector is the most accurate?

No detector publishes accuracy that holds for every kind of text. Pick from the top of the ranking, then run the 20-text test on your own writing. Read more

Is a free detector good enough?

For a first look, yes. GPTZero and Sapling have free tiers that renew. Before acting on a score, check the tool's published false-positive rate. Read more

Can I accuse someone based on a detector score?

No. A score is evidence to discuss, not proof. Detectors flag human writing, most often from non-native English writers. Read more

Which one should a school, an editor or an agency use?

The ranking re-sorted for each use, with a pick and the reasons. Read more

Prices and plans change

Get an email when the ranking changes