Sentence length variation
—
Low variation is common in AI text
Vocabulary diversity
—
Unique words ÷ total words
Avg sentence length
—
Words per sentence
Common transition words
—
"Furthermore", "Moreover", etc.
Why AI detection isn't reliable
No tool, including expensive commercial products, can definitively prove whether text was written by AI. These tools work by measuring statistical patterns — but human writers, especially skilled or non-native English writers, frequently produce text with the same "uniform" patterns these tools flag.
Academic institutions and publications have increasingly moved away from relying on AI detectors for disciplinary decisions because of high false-positive rates. Use this tool only as a casual, informal reference point — never as definitive proof.
Frequently asked questions
Can this prove text was written by AI?
No. This and all similar tools measure surface patterns, not authorship. False positives (flagging human writing as AI) and false negatives (missing actual AI text) are both common and well-documented.
Why did my own writing get flagged?
Clear, well-structured writing — especially from careful or non-native English writers — often resembles patterns associated with AI text. This is a known limitation across all AI detection tools, not unique to this one.