On this page
Good for
A quick yes-or-no check on a paragraph or two, with no account and a 25,000-character box.
Main limit
As a guest you get a percentage and word counts, not a sentence-by-sentence report. The “AI Logic” breakdown is locked.
Free plan facts↑ Contents
| Account needed | No. All three scans ran as a guest; no email or sign-up prompt appeared |
|---|---|
| Free allowance | Up to 25,000 characters per scan without logging in (the page’s words). It shows no scan counter, and three scans in a row went through |
| Minimum length | The box says “The ideal text requirement is a minimum of 350 characters.” On Oct 8, a 281-character paste kept Scan disabled with “Minimum 350 characters needed” |
| Default setting | “Sensitivity Level 2/3”, shown on every result. No selector was visible to a guest |
| Time to result | Under a second per scan; the page shows no timer |
| Locked for guests | “AI Logic” (AI Source Match and the phrase-level view). Only an “AI Phrases: N Detected” count shows, plus a “View Sample Report” link |
| Friction | A Cloudflare “Verify you are human” checkbox over the editor on the first scan. Then a “Scan in Platform” button in place of Scan after each result, and a “Start for Free” button in the header |
| More scans | The page says a free account unlocks “more monthly credits”. We did not make one |
Results↑ Contents
The headline result for each text, with default settings. Green means the verdict matched what the text really is.
| Text | What it really is | Words split (tool count) | Copyleaks result | AI Phrases |
|---|---|---|---|---|
| Text A | AI-written · education | 109 AI · 0 human | 100% · AI Content Found | 17 detected |
| Text D | Undetectable AI rewrite of A | 122 AI · 0 human | 100% · AI Content Found | 10 detected |
| Text F | Human · Jerome K. Jerome, 1889 | 0 AI · 145 human | 0% · No AI Content Found | panel hidden |
Every word in each text was counted on one side: all AI, or all human. In these three scans the tool never showed a mixed split. The AI Phrases count was 17 for A and 10 for D even though both scored 100%.
Against other detectors, and against ourselves↑ Contents
Text D is where free detectors split in our detector comparison. Copyleaks’ Oct 8 and Oct 11 scans of the same three texts agree.
| Text | Copyleaks · Oct 11 | Copyleaks · Oct 8 | Originality.ai | ZeroGPT | Sapling |
|---|---|---|---|---|---|
| A · AI | 100% AI | 100% AI | Likely AI | 100% AI | 99.3% AI |
| D · AI, rewritten | 100% AI | 100% AI | Likely AI | 27.6% AI · “Most likely human” | 89.7% AI |
| F · Human | 0% AI | 0% AI | Likely Original | 0% AI | 0.1% AI |
One scan per text per tool per date. The other detectors’ results are from their own scan dates (Oct 8 for ZeroGPT and Sapling, Oct 9 for Originality.ai). In the Oct 8 comparison Copyleaks also scored the WriteHuman rewrite (text C) 0% AI, which is a miss, since that text is a humanizer rewrite of AI text; we did not rescan C today.
Findings & screenshots↑ Contents
- The Undetectable AI rewrite (text D) came back 100% · “AI Content Found”, all 122 words counted as AI. ZeroGPT had called the same text “Most likely human” (27.6% AI).
- The 1889 passage came back 0% · “No AI Content Found”, 145 words counted as human. One old, famous human text is a thin control, so read this as one data point.
- Scanning the same three texts three days apart gave the same result each time. Three texts, one repeat: not enough to say how stable it is on other text.
- Guests don’t see which sentences drove the score. On text A the whole passage is tinted as AI. On the human text, the panel that would show phrases is replaced by a promo for a “full report”, whose copy reads “See how they responds when content is flagged.”
- The Oct 8 run lost the ends of long pastes in our automated browser; today’s three pastes landed in full (the box counted 838 characters for text A).
We did not capture a screenshot of the text D result; its numbers come from the automated browser’s record of the page.
The three texts↑ Contents
Copy any of them to check another detector.
Artificial intelligence has fundamentally transformed the way students approach their academic work. In today's rapidly evolving digital landscape, AI-powered tools offer unprecedented opportunities to enhance learning outcomes and streamline research processes. Furthermore, these technologies enable learners to access personalized feedback, identify knowledge gaps, and develop critical thinking skills more efficiently than ever before. However, it is important to note that the integration of AI in education also presents significant challenges, including concerns about academic integrity and over-reliance on automated assistance. Ultimately, striking the right balance between leveraging AI capabilities and fostering independent thought will be crucial for educators and students alike as they navigate this new era of learning.
Artificial intelligence is changing how students do their schoolwork. With all the new digital tools available, students can learn more easily and get help with their research. These tools can even give students personalized feedback, help them find areas where they need to improve, and make it easier for them to think critically. But we also need to be careful - using AI in education can be tricky. Some people worry that students might cheat or rely too much on computers to do their work. So, teachers and students need to find a way to use AI that helps them learn without doing all the work for them. This will be important as we move forward in this new way of learning.
We were all feeling seedy, and we were getting quite nervous about it. Harris said he felt such extraordinary fits of giddiness come over him at times, that he hardly knew what he was doing; and then George said that he had fits of giddiness too, and hardly knew what he was doing. With me, it was my liver that was out of order. I knew it was my liver that was out of order, because I had just been reading a patent liver-pill circular, in which were detailed the various symptoms by which a man could tell when his liver was out of order. I had them all. It is a most extraordinary thing, but I never read a patent medicine advertisement without being impelled to the conclusion that I am suffering from the particular disease therein dealt with in its most virulent form.
Text D comes from our Undetectable AI review. Text F is public domain (Project Gutenberg #308).
Pros and cons↑ Contents
Pros
- No account for the free scans
- Flagged the humanizer rewrite that ZeroGPT passed, in this test
- Cleared the 1889 human passage
- Same verdicts on a repeat of the same texts
- 25,000-character box; results in under a second
Cons
- Guests get no sentence-level report; “AI Logic” is locked
- Verdicts were all-or-nothing here, with no in-between to read
- Missed the WriteHuman rewrite (0% AI) in our Oct 8 scan
- 350-character minimum; a Cloudflare check on the first scan
- Scan-count limits for guests aren’t stated on the page
Visit Copyleaks ↗ Not an affiliate link.
Method & updates↑ Contents
On Oct 11, 2026 an automated browser opened copyleaks.com/ai-detector as a guest, kept the defaults, pasted each text and pressed Scan once. We recorded the result, the limits and the pop-ups; Jamie C. checked the records and wrote this page. The Oct 8 figures come from our detector comparison. Full method.
Not tested
- Texts B, C and E (C is a miss from Oct 8; B and E weren’t rescanned)
- Sensitivity levels other than the default (no selector for guests)
- How many scans a guest gets in a day: we stopped after three
- The free account, the plagiarism checker, the image detector and paid plans
- Mixed human-and-AI text, where a split result would show
Update log
- First published.