Method
Ranking criteria and test method
Five criteria anyone can check on a vendor's own site, a fixed formula, and text sets to test accuracy yourself.
| Criterion | Weight | 0 | 1 | 2 |
|---|---|---|---|---|
| Published evidenceDoes the vendor publish how it tests accuracy, including its false-positive rate? | ×2 | 0No accuracy claim, or none you can trace | 1A headline accuracy claim without a method | 2A published method or paper with a false-positive rate |
| Explains its scoreDoes it show which passages drove the score, or only a single number? | ×1 | 0A single score | 1A score plus a general label | 2Sentence or passage highlighting |
| Free to testCan you run your own test before you pay? | ×1 | 0No free access for individuals | 1A trial or one-time credits | 2A free tier that renews |
| Public priceIs the price on a public page, or behind a sales call? | ×1 | 0Sales or institution only | 1Partly public | 2Plans and prices on a public page |
| Fits a workflowCan it run where the work is done: an API, an LMS, a docs plugin, bulk scans? | ×1 | 0Paste-in box only | 1One integration or an API | 2An API plus team, bulk or LMS options |
02 · Text sets
What to test with
Same topics across sets, 300–800 words each.
03 · Steps
Reproduce the ranking
Fix the list
Detectors with a web product you can buy or access that returns an AI score for English text.
Included on the review date: 8 detectors. Excluded: research demos and tools without a public product page.Score each criterion from public pages
Read the vendor's own pricing, product and research pages. Record each score as 0, 1 or 2 with the page it came from.
Source of truth: the vendor's site on the review date. Each profile lists the pages read.Weight and total
Evidence counts double; the rest count once. The most a detector can score is 12.
total = 2·evidence + explains + free + price + workflowBreak ties the same way every time
More evidence first, then freer access, then alphabetical order.
sort by total desc, evidence desc, free desc, name ascTest accuracy on text sets, separately
Accuracy is never folded into the rank: it shifts with each model update and each kind of writing. Run the sets below on your own text.
Protocol: /test-your-detectorRe-review
Prices, plans and published claims are re-checked at every review, and the date at the top of the ranking changes with it.
Last reviewed: 19 November 2026
04 · Scores on 19 November 2026
Every score, with its source
| # | Detector | Evidence×2 | Explains | Free test | Price | Workflow | Total | Cost |
|---|---|---|---|---|---|---|---|---|
| 01 | CopyleaksSchools | Published testing method with false-positive rates | Highlights AI-written phrases, sentence by sentence | Free checks, 25,000 characters without login | Personal and Pro priced on copyleaks.com/pricing | Docs add-on, extension, MCP; API and LMS by quote | 12/12 | Personal $16.99/month ($13.99 billed yearly). Pro $99.99/month ($74.99 billed yearly). Education and Enterprise by quote. Check |
| 02 | GPTZeroTeachers | Publishes benchmarks with its false-positive rate | Sentence-level highlighting | Free plan, 10,000 words a month | Plans priced on gptzero.me/pricing | API, Chrome extension, team plans, LMS | 12/12 | Free: 10,000 words a month. Premium $12.99/month, Professional $24.99/month (billed yearly). Check |
| 03 | Originality.aiSEO publishers | Publishes accuracy data and false-positive rates | Sentence-level precision on every plan | 10,000 words, one time | All plans priced on originality.ai/pricing | Bulk scan, team seats, API on Enterprise | 11/12 | Free: 10,000 words, one time. Pro $14.95/month ($12.95 billed yearly). Enterprise $179/month. Check |
| 04 | PangramEditors | Model card and stated false-positive rate | Highlights the sections it reads as AI | 7-day free trial | Plans and API rates on pangram.com/pricing | API, MCP, Docs and Gmail, team plans | 11/12 | Individual $20/month after a 7-day free trial. Professional $65/month. API $0.05 per 100 words. Check |
| 05 | SaplingWriters | States false positives occur; no published rate | Scores each sentence | Free, 5,000 characters per check | Plans priced on sapling.ai/pricing | Metered API, browser extension, team plan | 10/12 | Free: checks up to 5,000 characters each. Pro $25/month ($12 billed yearly). API metered, $5 minimum. Check |
| 06 | ZeroGPTQuick checks | Accuracy claims; no published false-positive rate | Sentence highlighting | Free tier, $0 | Plans priced on zerogpt.com/pricing | Pay-as-you-go API | 9/12 | Free tier, $0. Personal plans from $14.99/month. API $0.042 per 1,000 words. Check |
| 07 | Winston AIScans & OCR | Headline accuracy claims | Prediction map, sentence by sentence | 14-day trial, 2,000 credits | Plans on gowinston.ai/pricing | API, team members, PDF reports | 9/12 | Free 14-day trial (2,000 credits). Essential listed at $18/month billed monthly; promotions change it. Check |
| 08 | TurnitinUniversities | Publishes its false-positive rate and how it holds it | Highlights text in the report | No individual access | No public price | Moodle, Blackboard, Canvas and other LMSs | 8/12 | Sold to institutions. No price on its public pages: ask your institution or Turnitin. Check |