Pattern Pressure: what it measures and what it cannot prove.
Detect cliché AI-slop scaffolds, unsupported boosters, repeated em-dash rhythm, and formulaic phrasing pressure without pretending burstiness proves authorship.
Plain-English method
Pattern pressure replaces naive burstiness claims with visible lexical markers. It is cheap, explainable, and useful for draft cleanup — not proof that a model wrote the text.
Mechanism and scoring
The score is a deterministic density/weighting pass over phrase dictionaries, boosters, and punctuation rhythm markers. It is intentionally transparent and easy to audit.
What this catches
- Generic AI-slop phrases
- Boosters and intensifiers
- Dense em-dash rhythm
- Formulaic marketing scaffolds
What this misses
- AI text that avoids clichés
- Human marketing copy that uses the same language
- Truth, source support, or authorship
This is one signal in a layered stack.
Single-method detectors are too easy to overtrust. Pattern Pressure is useful when it changes routing: allow, revise, human_review, or reject. It should be layered with specificity, provenance, pattern pressure, Unicode sanitation, media provenance, and paid Deep Scan when the decision matters.
check_pattern_pressure is available through local MCP with no LLM cost, and through the remote MCP endpoint with free unauthenticated rate limits. See all detection methodologies and the dedicated methodology page for the deepest treatment of this signal.