GPTZero and false positives

A widely used AI text detector with a free tier, popular with teachers and editors.

What it claims

GPTZero's own FAQ states its results should not be used to punish students and recommends its writing reports be treated as one input in a holistic assessment.

Where it fails

Short texts and general, essay-like prose are the documented weak points, and both are common in real work. Its "mixed" verdict is easy to misread as a finding when it is a hedge.

How to respond

Quote the company's own guidance back. It is far more persuasive coming from the vendor than from you.

Worth knowing

A "mixed" or partial result is not evidence of partial AI use. It is the model expressing uncertainty.