The Consistency score measures agreement across AI models in an AI Credibility analysis. High consistency means models tell a similar story; low consistency flags unstable or contested AI perception.
Consistency and model divergence are two sides of the same signal. Use them when deciding whether to act on a dimension score or gather more evidence first.
Large swings in Consistency between scan dates may indicate a repositioning taking hold, a PR event entering training corpora, or model updates changing recall.