AI Detector
Find out whether AI wrote the text with a detailed breakdown
Paste your text and get a probability score that artificial intelligence wrote it — including highlighted suspicious sentences and a breakdown across five metrics.
- A 0–100 % score and a clear verdict
- Suspicious sentences highlighted with a reason
- 5 contexts: general, academic, marketing, creative, social media
- 23 languages including English and other European languages
What the detector can do
More than a score — a full text analysis
Instead of a bare number, you get an explanation of why the text feels human or artificial.
Transparent analysis
You get an analysis, not just a number
A 0–100 % score is only the beginning. The detector highlights suspicious sentences, explains why and breaks the result down into five measurable factors.
Treat it as a strong indicator, not as absolute proof — the decision is yours, but with full information at hand.
- A 0–100 % score and verdict
- Highlighted suspicious sentences
- A reason for each passage
- Five detailed metrics
- Human vs. AI signals
- A clear report
Probably AI
Text type: Academic
Five contexts
Adapts to the type of text
An essay, a marketing copy and a social post all follow different rules. Pick a context and the detection fine-tunes its criteria precisely for it.
- General, academic, marketing
- Creative and social media
- More accurate results
Languages
23 languages including English
Detection isn't limited to English alone. You can check text in English, Czech, Slovak and other European languages.
- 23 languages
- English and other European languages
- More European languages
How it works
An analysis in three steps
- 01
Paste your text
Paste 50 to 10 000 characters of the text you want to check.
- 02
Choose the text type
Select a context — general, academic, marketing, creative or social media.
- 03
Read the analysis
You get a score, a verdict, highlighted suspicious passages and a breakdown across five metrics.
Who is AI Detector for?
Anywhere the origin of a text and its credibility matter.
Education
Teachers and tutors quickly check whether work was done independently or with help from AI.
View solutionMedia and authors
Verify the origin of contributions and source material from freelancers before you publish them.
View solutionMarketing
Check texts from suppliers and keep your content consistent and human-sounding.
View solutionFrequently asked questions about the detector
How long a text can I check?
From 50 to 10 000 characters per analysis.
Does the detector work in your language?
Yes. It supports 23 languages including English and other European languages.
What does the resulting score mean?
A 0–100 % score expresses the probability that AI wrote the text. It is accompanied by a verdict and a breakdown across five metrics. Treat it as a strong indicator, not as absolute proof.
Which five metrics does the detector track?
The score comes from a weighted combination of five indicators: Vocabulary (the density of typically “AI” words and phrases), Structure (artificial constructions such as vague references to “experts” or filler like “plays a key role”), Predictability (how easily the text can be guessed), Sentence variety (people vary sentence length, AI keeps it even) and Vocabulary diversity. Each metric carries a different weight in the overall score.
How should I read the verdict?
Below the score you'll find a worded verdict: roughly up to 35 % “likely human”, 35–65 % “ambiguous / mixed”, 65–80 % “probably AI” and above 80 % “very probably AI”. It is always a probability, not proof.
Does it spot specific suspicious sentences?
Yes. It highlights each suspicious passage directly in the text, labels it with high, medium or low severity and adds a short explanation of why it feels artificial. Alongside that, it also lists the human signals it found in the text.
Can I tell the detector what type of text it is?
Yes. You pick one of five contexts – general, academic, marketing, creative or social media text – and the detector adjusts its criteria accordingly. That way a formal academic style isn't automatically flagged as AI, and for social posts it treats typos and a casual tone as a human signal.
Which plan includes the AI detector?
The AI detector is part of the paid subscription from the Basic plan upwards. You'll find a comparison of plans and prices on the Pricing page.
Is my text safe?
Yes. We process your text in line with GDPR, we don't share it with third parties and we don't use it to train models.
Something missing? Get in touch.
Discover the whole GuideGlare platform
Besides the detector, one account also gives you chat, images, audio, documents and a translator.