Paste an article, a post, a review or a text from an external writer and see which sentences read as machine-made. The demo below checks a section of up to 3,000 characters without signing in; with a free account you see all seven detectors side by side.
Six more detectors
Search engines do not penalise text for being written with the help of AI; they penalise content with no value of its own. A high score on an article that was meant to draw on first-hand experience or an interview is therefore a warning about the content, not about the technology.
The check is most useful where you do not write the text yourself: with external writers, agencies or guest articles. It shows whether you got what you commissioned.
With an account you pick the text type before checking — blog, post, review or e-mail. Some of the detectors judge the text differently depending on the type, because a short post has a different rhythm from an article.
In an e-mail the check automatically leaves out the greeting, the polite closing lines, the signature and the quoted previous message, so it judges only what the author actually wrote.
With the Pro plan you upload several files at once and each one is evaluated separately — useful for an editorial team that receives texts from several writers, or for checking a whole set of articles.
The results are saved to your history. When you check the same or an edited text again, each number also shows the change since the previous check.
No AI text detector is one hundred per cent accurate, and short texts such as posts or reviews are less reliable for it. One high number is therefore not enough to send a text back to its author.
It is better to agree on the rules in advance — when and how the writer may use AI — and to treat the result of a check as a reason to ask a question, not to make an accusation.
Search engines judge whether content is useful, not who wrote it. The problem is mass-generated content with no value of its own, and the check helps you catch it before you publish it.
At least 100 characters, but for a short post the result is indicative only. Checking the whole article is more reliable.
You can agree that texts will go through a check, but don't put the score into the contract as the only criterion. It is more reliable to agree on how the writer may use AI and to treat the result as a basis for a conversation.
Sometimes yes, sometimes no. Humanisation changes exactly the properties some of the detectors look at, which is why we show several detectors side by side, together with the marked sentences.
Texts up to 300,000 characters, seven detectors at once, colour-marked sentences and a history of your checks. The Pro plan adds file upload and checking several texts at once.