Model divergence (or cross-model divergence) describes how far apart ChatGPT, Claude, Gemini, Mistral, Perplexity, and other models score or describe the same brand.
Low divergence suggests stable, shared AI knowledge. High divergence may indicate thin training data, recent brand change, ambiguous positioning, or category confusion — warranting deeper review of model reasoning and drivers.
AIRepScore surfaces divergence so you do not over-trust a single model's story.