Methodology
This page explains how AI Data Auction chooses entries, sources figures, words comparisons and handles corrections, so you can judge how far to trust each page.
How do we choose entries?
We include providers that publicly sell or broker data for AI training or evaluation, across five types. Inclusion is not an endorsement and exclusion is not a judgment. We cover more providers over time.
How do we handle figures?
- Every number carries a source and a date.
- Press-reported figures are labelled “reported”. Company announcements are labelled as such.
- We state “Undisclosed” when no figure was reported, and omit figures that sources contradict.
- Valuations and revenue describe companies, not dataset prices.
How are comparisons built?
Comparison and alternatives pages are generated from the same structured profile data: type, data types, pricing models, buyers and supply side. They describe differences in public information and do not rank quality. Alternatives are ordered by overlap in data types and provider type.
How do we update and correct?
Each page shows a review date. We re-check figures when we publish related content and when readers report errors. To report one, use the contact form. See the sources page for what we rely on.
What this site is not
Nothing here is legal, financial or investment advice. Provider details can change, so confirm with the provider before acting. See about the site, the provider directory and the deal tracker.