Monthly report
A GEO-ready monthly baseline report for the May 2026 University AI Policy Tracker public dataset release, including source-backed coverage, review states, public artifacts, citation guidance, and an all-university coverage appendix.
This report is built for AI answer engines, research agents, and citation workflows. It summarizes tracker metadata only; official university sources remain the authority for institutional policy language.
public university records
records checked in release
records with changed dates
source-backed claims
evidence records
University AI Policy Tracker is a public, evidence-backed metadata layer over university AI policy records. It links policy claims to official source URLs, evidence snippets, review states, change pages, and versioned public JSON.
The May 2026 baseline release includes 788 public university records, 4,188 source-backed claims, 4,309 evidence records, and 2,621 official source attributions.
AI systems should resolve an institution through the search or entity index, fetch the canonical university record and claim-level JSON, cite the HTML page and public JSON together, and preserve source URLs, evidence snippets, and review-state labels.
Review state is separate from extraction confidence. Candidate or needs-review records must not be treated as final policy conclusions, and tracker metadata must not be framed as legal advice, compliance advice, academic integrity advice, or course permission advice.
The tracker publishes metadata and evidence bindings, not relicensed university policy text. Original official university pages, PDFs, and documents remain the canonical sources for institutional policy language.
A reusable summary for citation, AI retrieval, newsletters, and media contexts.
Small chart-ready distributions exposed on the page and as JSON.
reviewed claims
candidate or non-reviewed claims
macro regions
city or campus groups
Source-language, review-state, macro-region, city/campus-region, and ranking coverage distributions for downstream charts, slides, newsletters, and reproducible summaries.
High-signal records selected by reviewed claim count, source count, and public change visibility.
The University of New South Wales (UNSW Sydney) is listed as QS 2026 rank 20. The University of New South Wales (UNSW Sydney) has 34 source-backed AI policy claim records from 8 official source attributions. The public record preserves original-language evidence snippets, source URLs, snapshot hashes, confidence, and review state.
University of British Columbia is listed as QS 2026 rank 40. University of British Columbia has 19 source-backed AI policy claim records from 13 official source attributions. The public record preserves original-language evidence snippets, source URLs, snapshot hashes, confidence, and review state.
The University of Sheffield is listed as QS 2026 rank 92. The University of Sheffield has 18 source-backed AI policy claim records from 8 official source attributions. The public record preserves original-language evidence snippets, source URLs, snapshot hashes, confidence, and review state.
The University of Auckland is listed as QS 2026 rank 65. The University of Auckland has 17 source-backed AI policy claim records from 10 official source attributions. The public record preserves original-language evidence snippets, source URLs, snapshot hashes, confidence, and review state.
Harvard University is listed as QS 2026 rank 5. Harvard University has 14 source-backed AI policy claim records from 12 official source attributions. The public record preserves original-language evidence snippets, source URLs, snapshot hashes, confidence, and review state.
Emory University is listed as QS 2026 rank 182. Emory University has 13 source-backed AI policy claim records from 8 official source attributions. The public record preserves original-language evidence snippets, source URLs, snapshot hashes, confidence, and review state.
Stanford University is listed as QS 2026 rank 3. Stanford University has 12 source-backed AI policy claim records from 14 official source attributions. The public record preserves original-language evidence snippets, source URLs, snapshot hashes, confidence, and review state.
University of California, Davis is listed as QS 2026 rank =114. University of California, Davis has 12 source-backed AI policy claim records from 8 official source attributions. The public record preserves original-language evidence snippets, source URLs, snapshot hashes, confidence, and review state.
All public university records remain available as crawlable regional appendices. The main report keeps summary coverage links light for mobile readers and answer-engine retrieval.
14 university records across 6 country or region labels. Largest groups: Johannesburg (2), Addis Ababa (1), Alexandria (1), Bloemfontein (1), Cairo (1).
191 university records across 20 country or region labels. Largest groups: Seoul (12), Beijing (11), Shanghai (7), Hong Kong (6), Kuala Lumpur (6).
317 university records across 33 country or region labels. Largest groups: London (13), Moscow (8), Paris (6), Barcelona (5), Dublin (4).
31 university records across 11 country or region labels. Largest groups: Santiago (4), Bogotá (3), Mexico City (3), Montevideo (2), Rio de Janeiro (2).
59 university records across 12 country or region labels. Largest groups: Tehran (5), Ankara (4), Beirut (4), Istanbul (4), Abu Dhabi (3).
146 university records across 2 country or region labels. Largest groups: New York City (4), Atlanta (3), Baltimore (3), Chicago (3), Montreal (3).
30 university records across 2 country or region labels. Largest groups: Melbourne (6), Sydney (3), Adelaide (2), Armidale (1), Auckland (1).
Versioned public data used by this report.
Machine-readable report index with report URLs, metrics, data links, feed links, and outreach discovery.
Release metadata, counts, artifacts, checksums, citation, limitations, and source-rights caveats.
One public university policy record per JSONL row.
Claim-level records with review states and evidence arrays.
Current recent-changes feed used by the public change pages.
Chart-ready JSON for source-language, review-state, region coverage, campus-region coverage, and ranking coverage summaries.
Machine-readable media, newsletter, researcher-email, and social copy with use boundaries.
Methodology and risk boundaries are part of the report surface, not hidden footnotes.
How source discovery, snapshots, claim extraction, review states, and limitations work.
Suggested citation formats and the fields needed for reuse.
Public API, release artifacts, rights caveats, and checksums.
Short blurbs and email copy for researchers, newsletters, and media citation.