AiDetego

AIDetego is a free online tool for detecting AI generated texts. You paste a text and AIDetego analyzes whether it was likely written by an AI like ChatGPT, Claude or Gemini.

The tool is aimed at teachers, editors, recruiters and anyone who wants to know whether a text was written by a human or an AI.

The analysis shows you a confidence score and highlights suspicious text passages in color.

You paste the text to be checked into the input field. AIDetego analyzes the text based on several statistical features.

Perplexity measures how predictable the text is. AI generated texts typically have low perplexity because they prefer statistically likely word sequences. Human text is often more surprising and variable.

Burstiness describes the variation in sentence length. Humans naturally write both short and long sentences. AI texts tend to produce more uniform sentence lengths.

N gram distribution analyzes the frequency of word groups. Certain phrases and transitions appear more frequently in AI texts than in human text.

The result is displayed as a confidence score from 0% (probably human) to 100% (probably AI). Suspicious text passages are color marked so you can see exactly which parts the analysis classifies as AI typical.

AIDetego is used in various fields.

Teachers check submitted homework, essays and seminar papers. The score gives a first indication of whether a text was possibly created with AI assistance. It is important to note: AIDetego is an indicator, not proof. The results should serve as a starting point for a conversation, not as the sole basis for an assessment.

Editors and content managers use AIDetego to check submitted guest articles. High quality content is essential for SEO and reader trust.

HR departments check cover letters and motivation letters. A fully AI generated cover letter says something about the effort an applicant has invested.

Publishers and literary agencies use AIDetego as a first screening instance for submitted manuscripts.

AIDetego is built as a web app with Next.js and TypeScript. The analysis algorithms use statistical text analysis methods like perplexity, burstiness and n gram distributions.

Monetization is through Google AdSense.

The entered texts are only processed for the duration of the analysis and are not permanently stored. There is no registration and no user account.