Verification

How GPTypo verification works

A verified scan means a fine-tuned neural model evaluated the writing and returned a probability. It gives you something to show when another detector flags your writing.

Last updated: May 2, 2026 · Tested on 4,133 labeled samples
01

Every scan returns a score from 0 to 100

100 means the model is highly confident the writing is human. 0 means highly confident it is AI-generated. Most real writing lands somewhere on the spectrum, which is why we group results into three zones.

87 · Likely Human
0255075100
AI Detected
0 to 45
Mixed
46 to 74
Likely Human
75 to 100
02

Verified vs estimate, and why the difference matters

GPTypo runs two scoring systems with different reliability. Only one carries the ✦ Verified badge. The numbers below come from automated testing against 4,133 labeled samples.

✦ Verified · Pro

Neural verification

A fine-tuned classifier trained on a curated corpus of verified human and AI-generated text spanning 11 AI models, 8 domains, and 12 adversarial attack types. Independently evaluated on the RAID benchmark at 98.2%.

99.9%
AI flagged
2.9%
False positives
Estimate · Free

Local heuristic

A 5-signal scorer that runs in the extension itself. Useful for a quick read before submitting work, but the false positive rate is more than ten times higher than verified scans.

23.1%
AI flagged
31.1%
False positives

For shareable proof reports, only verified scans carry the badge. Estimate scans can also be shared, but the absence of the badge is itself a signal to the viewer.

03

Why verification exists at all

Across all major AI detectors, formal human writing is misclassified at high rates. Heuristic scoring approaches similar to what most free detectors use will misclassify human writing nearly a third of the time on a labeled corpus.

31.1%
of human writing flagged as AI by free-tier heuristic scorers like GPTypo's
measured against GPTypo's free-tier scorer on a labeled corpus of 1,268 human samples · see the full data

When a human writer gets caught in that error rate, they need something to point at. Verification is built to be that something. A timestamped, link-shareable scan from a separate model the accuser does not control.

04

What verification is not

Worth being clear about the limits, because they affect when and how a verified scan is useful.

i
Not a humanizer
GPTypo does not try to disguise AI-generated writing as human. The point is to verify writing that is already human, not to launder writing that is not.
ii
Not legal proof
A high score does not mean a court would accept your writing as human-authored. It means a separate model scored it that way, on a specific date, with the text shown.
iii
Probabilistic, not certain
No model claims certainty. A score of 87 means the model leans strongly toward human. It does not claim a literal probability of human authorship in some absolute sense.
iv
Not a replacement for the detector you are accused by
If a school or platform requires a specific detector, verification is a complement, not a substitute. It is the second opinion you point at when the first one is wrong.
See your writing's verified score
Free includes estimate scoring. Verification is included with Pro.
Try GPTypo free
Read next
How GPTypo works → the product, the workflow, the surfaces