Detect AI-Generated Text

Paste in text to have the perplexity and burstiness analyzed to see the probability that it was generated by AI.

The adversarial AI model in this chat detects AI-generated content by examining various factors, including perplexity and burstiness of the text. The model assigns a likelihood score ranging from 0 to 1, with lower scores indicating a higher probability that the text is human-written and higher scores suggesting it is AI-generated.



Perplexity is a measure of how well a probability distribution predicts a sample, and it is often used to assess the quality of language models. If a text has a low perplexity score, it means the language model accurately predicted the words in the text, suggesting that the content may have been generated by an AI.

Burstiness refers to the tendency of certain words or phrases to appear in clusters within a text. AI-generated content tends to have more repetitive patterns and higher burstiness compared to human-written text.

By analyzing these factors and any other relevant features, the model assigns a likelihood score to the given text. The explanation provided by the model will highlight the key factors it considered when determining whether the text was likely written by a human or generated by AI. If the score is high and you suspect plagiarism, it is advisable to run a separate test for plagiarized content, as this model is primarily focused on detecting AI-generated text.