Google position one gets roughly ten times more clicks than position ten. Every SEO team on the planet knows this number. It is printed in decks, cited in pitches, used to justify budgets. The entire discipline of search engine optimization exists because of it.
Nobody has told you the equivalent number for AI responses. Here it is: a brand mentioned in the first position of an AI-generated answer receives approximately five times more cognitive engagement than the same brand mentioned in the fourth position. Not five times more clicks. Five times more attention. The distinction matters more than it sounds.
Why clicks and attention are not the same thing
In Google search, position determines whether you get clicked. The user sees a list of results, scans titles and descriptions, makes a choice, and navigates away. The ranking affects traffic. Traffic is measurable. The loop is closed.
In an AI response, there is no list to click. There is a paragraph, sometimes several. The user reads, or partially reads, or stops reading when they have what they need. Your brand does not compete for a click. It competes for a moment of cognitive engagement inside a continuous stream of text. If that moment does not happen, the user does not register your brand at all, regardless of whether you technically appeared in the answer.
This is a fundamentally different competitive dynamic. And it is one that your current measurement stack almost certainly cannot see.
The intent gap
When someone types a query into Google, they are beginning a search. They have a question and they are looking for a starting point. They expect to click several times, compare options, read reviews, and eventually make a decision somewhere downstream.
When someone asks an AI a question, they are often ending a search. They want an answer, not a list of options to investigate further. The intent is higher, the patience is lower, and the tolerance for irrelevance is close to zero. A user who asks "which analytics platform should I use to measure brand visibility in AI" and receives a four-paragraph response is not going to open six tabs and comparison-shop. They are going to take the first credible answer and act on it.
This changes everything about what position means. In Google, position three still gets meaningful traffic. In an AI response, position three is often where attention has already begun its fastest decline. The decay curve for AI chat is steeper than for almost any other content format we have measured. It drops faster than social feeds, faster than web editorial, faster than email.
The citation asymmetry
There is a second dimension to AI positioning that has no direct equivalent in search: the citation layer.
When an AI model mentions your brand, it often cites a source. That source is not chosen randomly. Models weight citations toward domains with high content quality signals, consistent publishing patterns, and strong topical authority. If your brand appears in position one but the citation points to a competitor's domain as the source of information about you, the attention value of that mention is significantly lower than it appears.
Conversely, a brand that appears in position two but is cited directly from its own authoritative content may generate more decision-relevant attention than the brand ranked above it. Position and citation interact. Measuring one without the other gives you an incomplete picture.
What the data from 300 AI engines shows
BAX calibrates attention scoring across more than 300 AI engines and language model platforms. The positional decay patterns are consistent enough to model, but they vary meaningfully by query type, response length, and platform.
On conversational platforms with shorter response formats, the attention differential between position one and position two is relatively modest. On platforms that generate longer, more structured responses, the gap between position one and position three can be as large as the gap between page one and page two of Google search results.
This means that a brand optimizing for AI visibility without accounting for platform-specific decay is optimizing against itself on some channels while over-investing on others.
The practical implication
If you are running an AI visibility program today, you are almost certainly measuring share of voice: how often your brand appears across a set of queries, compared to competitors. This is the AI equivalent of tracking impressions. It is a starting point, not a strategy.
The next question, the one that connects AI presence to actual business outcomes, is: where in the response does your brand appear, in what context, cited from what sources, on which platforms, and how has that changed over the last 90 days?
That is the question the BAX Index answers. Share of voice tells you that you exist in the AI conversation. BAX tells you whether anyone is paying attention when you do.
One number to remember
Position one in an AI response. Everything else is a footnote.
Not because the other positions have no value. They do. But because the attention economics of AI-generated content are more concentrated at the top than any medium that preceded them. The user came for an answer. The first credible answer they find is usually the last one they need.
Your competitors' SEO teams figured out page one of Google twenty years ago. Your AI team has roughly eighteen months before this positioning dynamic becomes as well understood, as contested, and as expensive to win as search ranking.
The brands that move first will not just win position one. They will define what it means to be credible in their category when a machine is doing the recommending.