The vocabulary of trust in writing.
Plain-English definitions of the terms behind content authenticity, AI detection, and behavioral certification — with the honest caveats included.
AI detector false positive
An AI detector false positive is human-written text wrongly flagged as machine-generated. Documented rates are highest for non-native English speakers — one Stanford study found detectors flagged 61% of TOEFL essays as AI-written.
Read definitionBehavioral certification
Behavioral certification proves content origin by recording the creator's behavior during creation — typing rhythm, pauses, revisions — and issuing a signed certificate, instead of classifying the finished output.
Read definitionBurstiness
Burstiness is the variation in sentence length and structure across a text. Human writing tends to alternate short and long sentences unpredictably, while AI-generated text is often more uniform — so detectors use low burstiness as a machine signal.
Read definitionC2PA / Content Credentials
C2PA (Coalition for Content Provenance and Authenticity) is an open standard for cryptographically signed metadata that records how an image, video, or audio file was created and edited. Content Credentials is its consumer-facing label.
Read definitionContent authenticity
Content authenticity is the assurance that content is what it claims to be — created by the claimed author, through the claimed process, and unaltered since. In the AI era it has become a verifiable property rather than an assumption.
Read definitionContent provenance
Content provenance is the documented history of a piece of content: who created it, how, when, and what changed along the way. It answers authenticity questions with records instead of after-the-fact analysis.
Read definitionKeystroke dynamics
Keystroke dynamics is the analysis of typing rhythm — the timing between key presses, pause patterns, and bursts — which is distinctive enough to characterize how a specific piece of text was physically produced.
Read definitionPerplexity
Perplexity is a measure of how predictable a piece of text is to a language model. Low perplexity means the model finds each next word unsurprising; AI detectors treat low perplexity as a hint that text was machine-generated.
Read definitionProof of authorship
Proof of authorship is evidence that a specific party created a work. For writing in the AI era, the strongest proof documents the composition process itself, since finished text no longer carries reliable origin signals.
Read definitionWriting telemetry
Writing telemetry is metadata captured during composition — keystroke timing, pause durations, revision events, paste sizes, and session boundaries — recorded without storing the content itself.
Read definition