Grayness Score
A composite, normalized score indicating how genuinely contested a legal question is, built from eighteen tests across three layers (signals within single opinions, patterns across the corpus, and the structural stability of the doctrine); the source paper calls it the Gray Area Score.
First used in Detecting Genuine Doctrinal Ambiguity
The first layer reads individual opinions for signals such as explicit acknowledgment of a split, declarations of first impression, judicial hedging, dueling authorities, dissent vigor, unweighted balancing, canon conflict and precedent instability.
The second layer aggregates across cases: stance inversion, jurisdictional divergence, temporal drift without formal overruling, outcome variance, dissent rates and reversal rates. The third tests the doctrine's structure, including fragmentation into inconsistent sub-rules and different courts applying different tests to the same question, before the signals are combined into one normalized composite.
The score is meant to travel with a verifiable explanation that quotes the judicial language producing it, so the output is tied to extant text rather than to model probability. Its purpose is to move legal AI from outcome prediction toward contestation detection.