On ChatGPT API, running shoes: ASICS and HOKA sit on the top shelf, each rate published with a 95% interval; the cell is still measuring toward the precision gate.

Name: Veliu Luce: how AI engines recommend brands
Creator: Veliu Research Lab
License: https://creativecommons.org/licenses/by/4.0/

A single measured fact from the Luce board, with its interval, its sample, and where it was observed. Cite this page, and the number carries its own uncertainty.

Mention rate · 95% CI

ASICS83% 67-97

HOKA83% 70-97

Adidas23% 10-40low precision

Brooks67% 47-83low precision

New Balance57% 40-73low precision

Nike53% 37-70low precision

On Running7% 0-17low precision

Saucony63% 47-80low precision

Metric: probe-design mention rate over the Luce grammar; not demand-weighted
Sample: 30 prompts
Surface: ChatGPT API
Wave: luce-live-w30
Provenance: L1 elicited
Estimator: joint cluster bootstrap over prompt clusters, percentile interval at 95%

See this cell's prompts, one by one

Download the reproducibility bundle for this cell (what is in it)

Cite this

On ChatGPT API, running shoes: ASICS and HOKA sit on the top shelf, each rate published with a 95% interval; the cell is still measuring toward the precision gate. Veliu Research Lab, Luce wave luce-live-w30, methodology luce-live-v0, measured from 15 Jul 2026, n=30. https://luce.veliu.com/claim/luce-live-w30/running-shoes-chatgpt-api. Veliu sells services to brands measured here.