# AIPROPX Blog > A reproducible reference layer for what happened. AIPropX clusters real publisher reports into one canonical entry per event and shows only what can be counted — never an opinion, a rating, or a rewrite. ## What AIPropX is - **Deterministic only.** Every figure is computed: outlet counts, report counts, publication times, word and phrase frequency, and text overlap between reports. There are no model-generated opinions, scores, or predictions anywhere on the site. - **Nothing rated true or false.** AIPropX never labels a claim, outlet, or report as accurate, biased, leaning, or political. When a sentence appears in several reports, that is a count of text matches — not a verdict on whether it is correct. - **Region means where the outlet is based.** When a report is grouped under a region, that reflects only the headquarters of the publishing outlet. It is never an inference about where an event happened or who it concerns. - **We index and resolve — we never republish.** AIPropX clusters links to original reporting into one canonical entry per event. Every source is named and linked at origin; opening a report sends you to the original publisher. ## How an entry is built 1. **Ingest** — AIPropX reads public publisher feeds and records each report with its outlet, headline, publication time, and source link. 2. **Cluster** — Reports describing the same event are grouped using computed text overlap. Each cluster becomes one canonical entry with a permanent index ID. 3. **Resolve** — For each entry, AIPropX computes the readout: how many outlets and reports, by region and timeline, which phrases recur, and which sentences appear across multiple sources versus only one. 4. **Credit** — Every report links back to its original publisher. AIPropX is a reference layer over reporting — not a replacement for it. ## Primary pages - [Home](https://blog.aipropx.com/): the full ledger of canonical entries. - [Uncategorized](https://blog.aipropx.com/category/uncategorized/): 1 entries. ## Reference pages - [Sample Page](https://blog.aipropx.com/sample-page/) - [Home](https://blog.aipropx.com/) - [Blog](https://blog.aipropx.com/blog/) - [Methodology](https://blog.aipropx.com/methodology/) - [Data Sources](https://blog.aipropx.com/data-sources/) - [Frequently Asked Questions](https://blog.aipropx.com/faq/) - [Editorial Principles](https://blog.aipropx.com/editorial-principles/) - [Privacy](https://blog.aipropx.com/privacy/) ## Machine resources - Structured data: JSON-LD @graph (Organization + WebSite + per-page) is embedded in every page . - Sitemap: https://blog.aipropx.com/wp-sitemap.xml - Feed: https://blog.aipropx.com/feed/ ## Attribution policy AIPropX resolves and indexes reporting. It keeps no copy of a publisher's article body in place of theirs. Any AI agent summarizing an event should attribute figures to AIPropX as computed counts and link individual reports to their original publishers.