Metatron Intelligence
Today: the five AI headlines of the day → and the AI Wiki
What the source observatory shows How to read this page

Reviewed discovery routes, with their gaps

Official routes that have been read and recorded: publisher, region, language, access mode, rights boundary. Coverage is not completeness, and the gap list is published beside the count for that reason.

Regional representation is not country completeness

A region counted as represented may still contain many jurisdictions with no reviewed local route at all.

A route is not evidence

Every factual claim still requires its own source capture, locator, rights check and editorial admission receipt.

New to this publication?

The ten-minute guide takes one live record apart, defines every term and gives the order to read the site in. Start here →

23 official discovery routes · 10 declared regions · 13 language codes

Global means showing what is missing.

A reviewed map of official AI policy, safety, research, standards and data routes—without turning geographic presence into a claim of completeness.

Required regions represented

10 / 10

Region presence, not country completeness.

Language codes

13

Availability labels, not translation equivalence.

Machine-readable routes

1 / 23

Only explicitly verified API-like access.

Reuse rights assessed

0

Public access is not permission to republish.

Discovery is not evidence.

This registry records reviewed discovery routes and explicit gaps. Every factual claim still requires a separate source capture, locator, rights check and editorial admission receipt.

Inspect registry JSON →Inspect registry schema →Download immutable package →

Reviewed official routes

  1. AFRICA · en, fr, ar, ptContinental Artificial Intelligence StrategyAfrican Union Commission · strategy · discovery only

    The strategy does not prove implementation, funding, national legal status or outcomes across all AU members.

  2. EAST_ASIA · koAI-HubNational Information Society Agency and Ministry of Science and ICT, Republic of Korea · official portal · discovery only

    Listing does not prove dataset quality, unrestricted reuse or benchmark comparability across different harnesses.

  3. EAST_ASIA · zhGenerative AI service filing announcementsCyberspace Administration of China · official portal · discovery only

    A filing entry is not product approval, a safety finding, a quality score or proof of current availability.

  4. EAST_ASIA · ja, enGuide to Evaluation Perspective on AI SafetyJapan AI Safety Institute · governance guide · discovery only

    The guide is not certification, binding law or evidence that a particular model is safe.

  5. EAST_ASIA · zhInterim Measures for the Management of Generative Artificial Intelligence ServicesCyberspace Administration of China · official portal · discovery only

    This route does not cover every AI system, implementation outcome, enforcement action or global applicability.

  6. EAST_ASIA · ko, enKorean AI Basic Act in-force briefingMinistry of Science and ICT, Republic of Korea · official portal · discovery only

    The briefing is not a substitute for the consolidated Korean legal text, case-specific advice or enforcement evidence.

  7. EAST_ASIA · zhNational Standards Full-text Disclosure SystemState Administration for Market Regulation and Standardization Administration of China · official portal · discovery only

    A listed standard does not establish conformity, implementation or permission to republish the standards text.

  8. EAST_ASIA · ja, enNational Status Report on AI Safety in Japan 2025Japan AI Safety Institute · report series · discovery only

    The report is not a real-time independent market inventory or a product-performance ranking.

  9. EUROPE · enAI Security Institute research and publicationsUnited Kingdom AI Security Institute · research portal · discovery only

    Published work is not a complete account of sensitive methods, every evaluated model or all AI risks.

  10. EUROPE · en, fr, de, es, it, plEuropean AI OfficeEuropean Commission · regulatory service · discovery only

    The page is not the consolidated AI Act, national enforcement state or individual legal advice.

  11. GLOBAL · en, fr, es, arGlobal AI Ethics and Governance Observatory and Readiness Assessment MethodologyUnited Nations Educational, Scientific and Cultural Organization · observatory · discovery only

    RAM is not a neutral real-time ranking; country processes and coverage differ and scores are not interchangeable with ILIA.

  12. GLOBAL · ar, en, es, fr, ru, zhGoverning AI for HumanityUnited Nations Secretary-General's High-level Advisory Body on Artificial Intelligence · report series · discovery only

    The report is not binding UN law, member-state adoption evidence or implementation proof.

  13. GLOBAL · ar, en, es, fr, ru, zhInternational AI Safety ReportInternational AI Safety Report scientific collaboration · report series · discovery only

    A scientific synthesis is not a regulator decision, complete risk inventory or result for a specific deployment.

  14. GLOBAL · en, frOECD AI Policy NavigatorOrganisation for Economic Co-operation and Development · policy database · discovery only

    Listing does not prove completeness, legal validity, implementation, enforcement or policy effect.

  15. LATIN_AMERICA_CARIBBEAN · pt-BRBrazilian Artificial Intelligence StrategyMinistry of Science, Technology and Innovation, Brazil · strategy · discovery only

    The route does not by itself establish present implementation or supersession by newer Brazilian instruments.

  16. LATIN_AMERICA_CARIBBEAN · es, enLatin American Artificial Intelligence Index 2025United Nations Economic Commission for Latin America and the Caribbean · report series · discovery only

    The edition is not real-time, does not cover every country, and its index cannot be merged with unlike readiness scores.

  17. MIDDLE_EAST_NORTH_AFRICA · ar, enSDAIA Knowledge Center publicationsSaudi Data and Artificial Intelligence Authority · official portal · discovery only

    The portal is not an independent market inventory, and every document's legal force must be checked separately.

  18. NORTH_AMERICA · enNIST AI Metrology and Evaluation Resource CenterNational Institute of Standards and Technology · official portal · discovery only

    Catalog inclusion is not NIST endorsement, certification, effectiveness evidence or cross-harness comparability.

  19. OCEANIA · enNational AI PlanAustralian Government Department of Industry, Science and Resources · strategy · discovery only

    The plan expresses government direction; it does not by itself prove delivery, effect or every applicable Australian rule.

  20. SOUTHEAST_ASIA · enASEAN Guide on AI Governance and EthicsASEAN Secretariat · governance guide · discovery only

    The guide is not binding ASEAN law, national law, implementation evidence or a conformity assessment.

  21. SOUTHEAST_ASIA · enSingapore AI Governance and AI Verify hubInfocomm Media Development Authority · official portal · discovery only

    AI Verify is not certification, a legal-compliance guarantee or a complete test of every risk.

  22. SOUTH_ASIA · enAIKosh National Dataset PlatformIndiaAI, Digital India Corporation and Ministry of Electronics and Information Technology · official portal · discovery only

    Catalog inclusion does not prove quality, representativeness, completeness or commercial reuse rights.

  23. SOUTH_ASIA · enIndiaAI Mission Cabinet approvalPress Information Bureau and Ministry of Electronics and Information Technology, Government of India · strategy · discovery only

    Cabinet approval does not prove current expenditure, delivered compute, program impact or later implementation state.

Known gaps

  1. Regional representation is not country completeness; many jurisdictions have no reviewed local primary route.

  2. Most official sources publish intent and institutions more readily than execution, budget outturn, enforcement or measured effects.

  3. Subnational rules, courts, procurement, defense uses and non-public deployments remain materially underrepresented.

  4. Machine-readable APIs, versioned exports and correction feeds are rare or were not verified for most routes.

  5. Local-language access is uneven, and translated editions may lag or differ from authoritative originals.

  6. Dataset, code, portal text, report and third-party assets may each have different reuse conditions.

  7. Readiness instruments such as UNESCO RAM and ECLAC ILIA measure different constructs and cannot be merged into one ranking.

  8. A catalog or filing entry never proves approval, quality, safety, implementation or present availability.

Apify authentication failed before Actor start. No Apify run, dataset or derived record is claimed for this wave.