Is the detector really free, with no email?

Yes. It asks for neither an account nor an email, the result shows on screen, and you can use it as many times as you like.

Does a language model read my text?

No. The calculation is deterministic: two analyses of the same text give the same result, and your text is not kept.

Why is my text turned down?

Either it is too short, or it is pasted as a single block, or it is neither in French nor in English. In all three cases, the result would be reassuring for the wrong reason.

How is this different from the post scan?

The scan emails you a score for your post against the corpus, with the substance and the hook. The detector only looks at writing tics, and shows the result right away.

AI writing tic detector

Does your post sound like an AI?

Paste your post. We count the writing tics found mostly in generated text, compare them to LinkedIn posts written by hand, and the result shows right away.

Instant result. Your text is not stored.

How it works

Step 1
Paste your post
With its line breaks, the way it will show on LinkedIn.
Step 2
Compared to human posts
Each tic is counted, then placed against what the authors in the corpus do.
Step 3
The result on screen
The tics in excess, the lines concerned, and how to rewrite them.

Free AI writing tic detector for LinkedIn

This detector does not say whether a text was written by an AI: nobody can honestly claim that. It counts writing habits that come up far more often in generated text than among human authors, and compares them to 4,064 LinkedIn posts measured one by one. No language model is involved, the calculation is deterministic, and the result shows right away. Free, no account, no email.

By Maxence Vroilant, founder of Never Boring AI · Updated on October 12, 2026

Tics, not a verdict
A human author can string three antitheses together, and a generated text may contain none. The detector measures habits, with their real frequency among humans.
It is the accumulation that counts
One tic alone proves nothing: many human posts carry one. It is when they pile up that the text starts to sound manufactured.
The target is never zero
We only flag excess. Aiming for zero devices would move you away from human writing rather than closer.

How does it work, in detail?

Three questions, and how we answer each one. Open the one you care about.

In French, three devices: the antithesis "it is not X, it is Y", several lines in a row opening with the same word, and the lesson announced by a label followed by a colon. In English, the set is narrower, because the English corpus is smaller.

Each device is counted in your text, then compared to what the human posts in the corpus do. A device is "in excess" when your post carries more of it than the vast majority of them.

What we get asked
the most

The questions that keep coming up about the tic detector.

No, and no tool can honestly claim that. It counts writing habits that are frequent in generated text and tells you whether your post carries more of them than the human posts in the corpus.

Who makes this tool

Never Boring AI is an AI agent for LinkedIn, built by Maxence Vroilant. This tool is open: no account, no email, no limit. It reuses a building block of the app, where the agent writes your posts from your experience and keeps your content calendar.