Independent indexes of what AI models recommend.
AI Indexes measures what AI models say when a buyer asks which product to use. Each index records the products the models name, how often, in whose favor, and with what reservations, by one fixed method, every month. Four indexes, one instrument: each asks every buying question six ways, on behalf of three sizes of company, of the same fourteen models on the same day, judges every answer by one published method, and publishes the raw record behind every standing.
The scope, every month
Every month, across the four indexes publishing: fourteen models from twelve labs, asked the same questions six ways on behalf of three sizes of company, every answer read by a disclosed judge, every label traced to the sentence it came from, every standing its own page.
| Index · edition | functions | categories | products | vendors | pages | prompts | answers | judge labels |
|---|---|---|---|---|---|---|---|---|
| GTM AI IndexOctober 2026 Edition | 6 | 112 | 6,834 | 3,544 | 16,914 | 2,016 | 28,224 | 148,766 |
| HR AI IndexOctober 2026 Edition | 7 | 58 | 3,245 | 1,753 | 8,734 | 1,044 | 14,616 | 74,522 |
| IT AI IndexOctober 2026 Edition | 7 | 136 | 7,649 | 2,443 | 20,697 | 2,448 | 34,272 | 177,409 |
| Finance AI IndexOctober 2026 Edition | 7 | 57 | 3,169 | 1,567 | 8,777 | 1,026 | 14,364 | 71,909 |
| Every month | 27 | 363 | 18,679 | 7,998 | 55,122 | 6,534 | 91,476 | 472,606 |
Figures are the current edition of each index, published on its own site; the record behind them is sent on request. Categories count once per index; a product or vendor counts once in each index it is named in, and once in the totals.
One instrument
Each index uses the same instrument: six framings of every buying question, three company sizes, fourteen models, a judge that writes one label per product named with the sentence it read as evidence, and a calibration repeat that measures how much a standing moves with nothing changed. Only a move larger than that noise floor is reported as movement. The method, the judge rubric and the second-judge check are published on every index and are the same on all four.
Independence
No placement is for sale, no ratings are collected from anyone, and no index recommends a product. Each standing is derived from the raw record, every figure traced on its page to the answers behind it; the record itself is sent on request. The indexes do not claim the models can be steered, and a standing that moves between editions is reported as a measurement, not a result.
The publisher's interests are declared where they touch a standing. The publisher is a cofounder of Gane.ai, which is building an operating system for B2B go-to-market. Its product overlaps with two categories of the GTM index: marketing automation and campaign management, and email marketing. Both are in the index rather than excluded, and those pages carry the disclosure. No category of the IT, HR or Finance index contains a publisher interest.
To cite
Cite the index, not the publisher. Every edition of every index is published under CC BY 4.0.
GTM AI Recommendation Index, [edition], gtm-ai-index.com. Published under CC BY 4.0.
IT AI Recommendation Index, [edition], it-ai-index.com. Published under CC BY 4.0.
HR AI Recommendation Index, [edition], hr-ai-index.com. Published under CC BY 4.0.
Finance AI Recommendation Index, [edition], finance-ai-index.com. Published under CC BY 4.0.
The subscriber app
The four indexes read for one company: every category it is named in, across all four, what changed against the previous edition, who the models pick instead, and every answer behind every figure. Seven days free, then $99 a month, $49 for each colleague. See the app and its price.
The catalog
The category map and product table behind the four indexes, offered as data: 363 categories, 18,679 products with their companies and websites, 8,052 descriptions, refreshed every month. License the catalog.
Contact
Corrections, claims of a product page and press, for any index: index@ai-indexes.com.