Interpreting your score

Read this before acting on any number this tool gives you.

The score is a density, not a probability

0–100 measures how densely your text carries patterns that are typical of raw, unedited model output. It is not "the probability this was written by AI" and not "the percentage written by AI". Bands: low (<25, few tells), moderate (25–49), elevated (50–71), high (72+, reads like raw model output).

Check the coverage line first

Each signal refuses to judge when there isn't enough text ("9 of 11 signals had enough text to judge"). On short texts most signals abstain and the score is built on almost nothing — treat anything under ~300 words as weak evidence, whatever the number says.

Known false positives

Human writing scores elevated when it is: formal or academic (few contractions, uniform sentences), written by non-native English speakers (the most documented harm across all detection tools — independent tests have measured other tools falsely flagging non-native writers at rates above 60%), heavily edited or house-styled, or simply short. If you're contesting a score, open the per-signal breakdown: every claim this tool makes is inspectable, which is the point.

Known false negatives

A light paraphrase or human editing pass removes most surface tells. A low score means the text reads human — it cannot certify how it was made.

Never use this to accuse anyone

No text detector — free or paid — is reliable enough to be the sole basis for an accusation, a grade penalty, or an employment decision. Vendors advertising "99% accuracy" are routinely measured far lower by independent tests. Use tell density as one signal among many, keep a human in the loop, and give writers the breakdown so they can respond to specifics.