How Red Stet measures authorship process
Red Stet's verifier surfaces specific patterns from a writing session that 30+ years of behavioral-biometrics and writing-process research has characterized in hand-typed composition — and that Red Stet's own calibration work extends toward separating hand-typed sessions from mechanical insertion. It does not run a black-box detector. The literature provides the within-human distributions; the human-versus-machine thresholds are ours, calibrated on a small corpus and published openly.
Methodology
One page per signal, plus how thresholds get set, what the composite computes, what the integrity checks prove, and what the system cannot do.
12 pages ↓How to use Red Stet
Setting up classes, writing in the editor, marking work, exporting and importing. Step-by-step guides organized by role.
Open /help/ →Policy & posture
Privacy framing, security model, FERPA posture, and the specs a reviewer can check without us.
Open /policy/ →Methodology
Tap a chip to filter to the pages written for your seat at the table.
Start here
What this section is, and the two pages to read before any individual signal.
What this section is. One page per signal Red Stet uses, plus five methodology pages that explain how we set thresholds, what the composite computes, what the integrity checks prove, what the system cannot do, and how this approach differs from AI-output detectors. Every claim carries its citation, its confidence level, and its limitations.
Who it's for. Academic-integrity board members reviewing whether Red Stet's evidence is admissible in their process. IT directors evaluating the tool against vendor claims. Journalists writing about the AI-authorship landscape. Researchers who want to verify our reading of the field. Writers curious about what the recording layer actually captures.
What we claim. The composition fingerprint is evidence FOR human authorship when its signals are present. It is not a verdict, not court-admissible biometric proof, and not an AI detector.
Combined fingerprint & composite scoring
How the seven signals combine into one number. Weights, formula, verdict tiers, and why no individual signal is enough on its own.
Investigation thresholds & baselines
What "flagged" means numerically, how we set the <1% / 1–2% / 2–5% / >5% ladder, and the small calibration corpus our thresholds rest on.
The signals
One page each — research history, mechanical reason, key papers, confidence level, limitations, and how Red Stet weights it.
Keystroke cadence variance
The time between consecutive keystrokes, and how much it varies. The best-validated behavioral biometric — 45+ years of research from Gaines 1980 through modern continuous authentication.
Correction rate
Backspaces per 1,000 typed characters. Cognitive process model of writing (Hayes & Flower 1980): revision is intrinsic to human composition, observable as keystroke-level edits.
Paste patterns
Paste size, frequency, and source (internal cut-and-rearrange vs external clipboard). Working-memory chunking (Miller 1956) makes paragraph-size external pastes mechanically incompatible with in-flow composition.
Mouse path geometry
Curve complexity of cursor paths between rest points. Fitts's law (1954) + Pusara & Brodley 2004 onward — human motor control produces sub-movement curvature; automation produces straight lines.
Click patterns
Whether clicks fall on grid-aligned pixels or with sub-pixel motor noise. Modest individually; useful in combination with other signals. Bot-detection literature, less academic.
Thinking pauses
Long inter-event gaps. Schilperoord 1996 + 40+ years of cognitive psychology of writing — macro-pauses are the observable traces of cognitive composition work.
Method & proof
What the checks behind "Verified" establish, and the citations under everything above.
Integrity checks
What the four checks behind "Verified" prove (internal consistency), what they can't (authenticity without the signature layer), and what the recorder captures beyond the scored signals.
Complete bibliography
All cited research, organized by topic. Open-access links where stable URLs exist.
Limits & comparisons
Where the method stops, and what it is not the same thing as.
What this system cannot do
The typed-out-model-output adversary, short documents, non-Latin scripts, accessibility input, measurement-environment confounds, the aware adversary. Eleven gaps, named explicitly.
Red Stet vs AI-content detectors
We measure process; they measure output. Different signal, different failure modes, complementary when stacked. Liang 2023, Sadasivan 2023, Weber-Wulff 2023.
Nothing here yet
We haven't written a methodology page for this reader yet. Tell us what you need: [email protected]
The framing. Red Stet's product is "is this human-authored?", not "is this AI?" Every signal on every page is described as evidence FOR human authorship — when it's present, the document is consistent with hand-typed composition. When it's absent, the patterns are uncommon in hand-typed documents — never AI-detected.
That framing is locked at the product level (see the 2026-06-08 decision drops). The signals don't tell us whether a model wrote the text; they tell us whether the SHAPE of the writing process is consistent with hand-typed composition. A skilled adversary who types model output character-by-character will produce a clean composition fingerprint. We name that limitation explicitly on the Limitations page.
An integrity board, an editor, or a reviewer reading a Red Stet verification gets evidence to interpret, not a verdict to apply.
Found something wrong? Email [email protected]. Last reviewed: 2026-06-08.