The reference for AI search visibility.

When a buyer asks an AI assistant for the best tool in a category, which brands does it name? That single question increasingly decides who gets shortlisted — before a human ever visits a website. The LLM Visibility Index measures it, in the open, across every major engine.

Since 2024

ZivRank has been researching how large language models surface brands since 2024 — building the measurement methodology, the prompt corpus, and the scoring model that power this Index. The first public edition of the Index was released in 2026, but the underlying research program has run continuously for nearly two years. We have watched AI assistants move from novelty to default research tool, and built the instrument to measure what that shift does to brand visibility.

Independence

Rankings derive solely from observed engine outputs. No company can pay to appear in the Index or to improve its position. Where a brand is cited, we record it; where it is absent, we record that too. The Index does not accept advertising, sponsorship, or placement fees of any kind.

Who runs it

The Index is published and operated by ZivRank SAS (RCS Créteil, France), a firm specialised in AI search visibility. ZivRank funds and runs the research, and is itself subject to the same measurement as any other company in its categories. Editorial control over scores rests with the methodology, not with commercial interests.

How to read a score

Each brand carries an AI Visibility Score from 0 to 100 combining five dimensions. A high score means a brand is consistently named, prominently placed, and cited across multiple engines on independent sources. A low score means it is effectively invisible at the moment of the buyer's question. The full breakdown is on the methodology page.

Corrections

If you believe a score misrepresents your brand, we want to know. Scores reflect our model and are revised monthly; we review every substantiated correction request. Contact us at [email protected].