AI Models Favor Familiar Brands in Search, Study Finds

A recent study of almost 4,000 responses to 66 buyer prompts in the United States shows that generative AI models are three times more likely to search for brands they already know. The models directed 55.7% of their brand‑related queries toward familiar brands, compared with only 17.4% for brands outside their top ten. Overall, only 31% of fan‑out searches mentioned a brand, and when a brand was named, 63% of those queries involved one of the five most familiar brands.

The research covered ten sectors—including travel, automotive, finance, business software, education, food and restaurants, luxury, fitness, and fashion—where the share of familiar‑brand searches ranged from 41% to 82%, versus 9% to 23% for unfamiliar brands. An example highlighted that the Gemini model searched for “Lemon Squeezy” while answering a question about online payment providers, illustrating that AI can still surface lesser‑known brands in certain contexts. The authors conclude that strong brand memory gives established companies an advantage in AI‑driven search, though high‑quality content can help newer brands gain visibility.