How Large Language Models Source Brand Reputation Across Languages and Markets (opens in new tab)

When a large language model (LLM) answers a question about a company, it grounds the answer in retrieved web sources, and those sources decide what the model says. Most analysis of AI brand visibility looks at the answer text. This study looks one step earlier, at the citations. We merge three Rankfor.AI datasets covering 128 brands across 12 home markets and 13 languages, and analyse 167,551 URL-grounded citations (189,974 total attribution row...