Why a detector needs to know what kind of text it is reading
Every genre has its own norm. An e-mail is short and polite formulas belong in it, a term paper is supposed to have structure, a listing is supposed to have bullets. When a detector knows what kind of text it is reading, it compares the text with how such a text usually looks and stops punishing properties that belong to the genre.
Same property, different meaning
Repetitive sentence structure is normal in a manual and suspicious in an essay. Without knowing the type, both count the same.
That is why DetekceGPT asks for the text type before the check. It decides what the text is compared against.
What Type template measures
It shows how closely the text follows the outline of its genre: paragraphs of equal length, bullets, headings, a salutation and a sign-off.
A high value does not mean AI. It means the text runs to form — which you expect in some genres and not in others.
What it changes in practice
For short operational texts, read the score loosely; for a longer piece of authored work, more strictly.
The same text judged as a different genre gives a different result. Picking the type is part of the task, not decoration.
FAQ
What if my text fits no type?
Pick the nearest one. Getting an exact match matters less than not comparing the text against a completely different genre.
Does the type affect the other detectors?
Mostly Type template and Style deviation, because both compare the text against something. The others measure the text itself.
Do I have to pick the type every time?
It is worth it. It is one click, and without it the text is compared against a generic norm.
Try it on your own text
Paste a text or upload a file and look at the score and at the specific sentences that came out suspicious.
Run a detection