The Forensic AI Content Detector is a browser-based tool that estimates whether a given text was likely written by a human or generated by an AI. It analyzes lexical diversity (perplexity-like score), sentence-length variation (burstiness), and highlights words commonly overused in AI writing so you can make an informed judgment.

Why Use This Tool?

• Transparency: Quickly flag content that may need human verification or disclosure.
• Editing: Identify AI-style phrasing to rewrite for a more natural tone.
• Research: Compare human vs. model outputs for datasets or evaluations.

Key Features

• Perplexity-style score: Approximates lexical predictability; lower diversity often suggests AI.
• Burstiness: Detects uniform sentence length typical of many LLM outputs.
• AI word highlighter: Marks terms such as “furthermore”, “leverage”, “comprehensive” that appear frequently in AI text.
• No uploads: All analysis runs locally in your browser.

How it Works

The tool splits the text into words and sentences, computes simple statistics (word frequency, sentence-length variance), and compares the vocabulary against a curated list of AI-favored words. A combined score and verdict are shown; highlighted text can be copied for further editing.

Try the Forensic AI Content Detector

Paste any paragraph above and click Analyze to see the AI likelihood score and highlighted phrasing. The tool runs entirely in your browser—no data is sent to any server.