Cross-platform divergence is the gap between what different AI engines say and cite for the same prompt, measured across ChatGPT, Google AI Mode, AI Overviews and Perplexity. A brand can be cited as a source by one engine, named without citation by another, and absent from a third, all on the same day and the same question. Reading one engine tells you about one engine.
Why the engines disagree
Each engine retrieves from a different index with different freshness, weights sources differently, and fans a prompt out into different sub-queries. Google’s AI surfaces are fed by Google’s index; ChatGPT and Perplexity run their own retrieval crawlers and lean on other indexes. Retrieval sources that only one engine reaches produce answers only that engine gives.
Why it matters for measurement
Divergence is why a screenshot from one engine is an anecdote. Citate samples each prompt many times on each surface and reports presence per surface, so the divergence itself becomes a finding: which engine you have won, which you have not, and which sources explain the difference. Convergence matters too. When a site loses its place in Google’s index, Glenn Gabe’s case work shows the AI surfaces that depend on that index follow it down together.
The Federation Index
Citate summarizes divergence with a federation index, a measure of how much the cited sources for a prompt are shared across engines versus split between them. A high split means the engines disagree about who the authority is, which is usually an opening.