Scripli Glossary

DefinedTerm

A blockchain timestamp anchor is a cryptographic commitment to a piece of data recorded on a public blockchain at a specific time. Every Scripli HAC is anchored to a public blockchain at the time of issuance. This creates an independent, permanent record of when the certificate was issued that can be verified by anyone without relying on Scripli. If Scripli ever ceased to exist, anchored certificates could still be independently verified.

DefinedTerm

A false positive in AI detection occurs when a tool incorrectly classifies human-written text as AI-generated. AI detectors analyse statistical properties of text output — vocabulary, perplexity, sentence structure — and produce probabilistic scores. Human writing that has these statistical properties (formal academic writing, carefully edited prose, writing by non-native English speakers) is disproportionately likely to be flagged as AI-generated even when it is entirely human-authored.

DefinedTerm

A Human Authenticity Certificate (HAC) is a cryptographically signed, tamper-proof record issued by Scripli that proves a specific writing session was carried out by a real human. It contains session metadata, behavioural metrics, a cryptographic hash of the session data, and a blockchain timestamp anchor. Anyone can verify a HAC publicly — no account required. HACs are non-transferable and permanently bound to the session that produced them.

DefinedTerm

The Provenance Score is a per-session metric computed by Scripli that reflects the cognitive effort present in a writing session. A high Provenance Score indicates signs of genuine intellectual engagement. A low Provenance Score means the writer produced the work with ease and flow. It is a signal of authentic effort, not a quality judgement. The Provenance Score is displayed on the public verify page for each HAC.

DefinedTerm

Perplexity is a statistical measure of how "surprising" or unpredictable a sequence of words is to a language model. AI-generated text tends toward low perplexity (predictable word choices), which AI detectors use as a signal. The problem: formal human writing — academic essays, legal documents, technical reports — also has low perplexity. This is why formal writing styles are disproportionately flagged by perplexity-based AI detectors.

DefinedTerm

Writing DNA is a writer's personal behavioural fingerprint — a profile built up from multiple HAC sessions on Scripli. It reflects stable, unique characteristics of how a specific person writes, developed from patterns observed across multiple certified sessions. Writing DNA becomes more distinctive and accurate as the writer accumulates more sessions, and is used to establish cross-document consistency.

DefinedTerm

Writing provenance is the verifiable record of the authorship process behind a written document — who wrote it, when, and how. Scripli defines human writing provenance as a behavioural + cryptographic proof chain that establishes a document was produced by a specific human in a specific session, independent of any analysis of the document's content. This is distinct from plagiarism detection (which checks sources) and AI detection (which analyses output).