Talk to the Institute

Alpha Institute • Responsible AI Observatory

Standards inform governance. Scope determines applicability.

An Observatory reading guide to primary sources, with clear separation between law, standards and recommendations.

Talk to Alpha

Read the original source

Check the issuing organization, edition, publication date, jurisdiction and current amendments. Licensed standards require appropriate access; a public summary is not a substitute for the full instrument.

Review the mapping

Control mappings require named reviewers, evidence expectations and an applicability decision. Crosswalks remain research candidates until approved. No one-to-one equivalence is assumed.

Track change

A changed source requires a new version and impact review. Preserve prior mappings and assessment versions rather than rewriting historical conclusions.

Standards reading library

Reading candidates, not approved methodology, verified legal interpretations or independent findings. Publication dates and versions marked unknown require review. No control mappings have been approved in this collection.

16 resources

NIST AI RMF

Organizational AI risk management.

NIST • Voluntary framework • United States

Version: 1.0 • Publication: 2023-01-26

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

NIST AI 600-1, Generative AI Profile

Risk management considerations for generative AI.

NIST • Voluntary guidance • United States

Version: AI 600-1 • Publication: 2024-07-26

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

ISO/IEC 42001

AI management systems. Restricted standard text; link and metadata only.

ISO / IEC • Management system standard • International

Version: 2023 • Publication: 2023-12

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

ISO/IEC 23894

AI risk management guidance. Restricted standard text; link and metadata only.

ISO / IEC • Risk management standard • International

Version: 2023 • Publication: 2023-02-06

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

OECD AI Principles

Trustworthy AI principles. Check the current recommendation and revisions.

OECD • Recommendation • International

Version: 2024 revision • Publication: 2024-05-03

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

UNESCO Recommendation on AI Ethics

Recommendation adopted on 23 November 2021. Ethical and human-rights considerations for AI governance.

UNESCO • Recommendation • International

Version: 2021 recommendation • Publication: 2021-11-23

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

EU AI Act

Applicability depends on role, system and current amended legal text. Legal review required.

European Union • Law • European Union

Version: Unknown • Publication: Unknown

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

Council of Europe Framework Convention on AI

Opened for signature on 5 September 2024. Party status, entry into force and domestic implementation require separate review.

Council of Europe • Treaty • Council of Europe

Version: CETS No. 225 • Publication: Unknown

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

FSB Responsible AI adoption practices

Twelve sound practices for responsible AI adoption by financial institutions. Final report following the June 2026 consultation.

Financial Stability Board • Voluntary guidance • Financial services

Version: Final report • Publication: 2026-10-15

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

Singapore Model AI Governance Framework

Organizational governance guidance; confirm applicable edition.

IMDA / PDPC • Voluntary framework • Singapore

Version: Unknown • Publication: Unknown

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

Singapore agentic AI framework

Updated agentic AI governance framework announced on 20 May 2026. Applicability and control mappings remain subject to review.

IMDA • Voluntary framework • Singapore

Version: May 2026 update • Publication: 2026-05-20

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

IEEE 7000

Ethical concerns in system design. Licensed text is not reproduced.

IEEE • Technical standard • International

Version: 2021 • Publication: 2021-09-15

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

OWASP LLM / GenAI Top 10

LLM application security risks. Exact publication date remains unconfirmed.

OWASP • Security guidance • International

Version: 2026 • Publication: Unknown

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

OWASP Top 10 for Agentic Applications

Agentic application security risks.

OWASP • Security guidance • International

Version: 2026 • Publication: 2025-12-09

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

MITRE ATLAS

Adversarial threats to AI systems. A living resource rather than a fixed certification standard.

MITRE • Threat knowledge base • International

Version: Unknown • Publication: Unknown

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

MIT AI Risk Repository

A research resource for describing AI risks, not a measure of incident prevalence.

MIT AI Risk Initiative • Research taxonomy • International

Version: Unknown • Publication: Unknown

Review: pending • Last reviewed: not yet reviewed • Review required

Metadata and original summary only; no full-text ingestion

Machine-readable resources

Alpha Evals

Explore existing readiness and independent evaluation pathways. https://evals.alpha.ac/

Explore

Choose the relevant path.

Responsible AI Observatory

Standards, research and board education connecting Responsible AI principles to accountable governance.

Explore

Research that makes governance examinable.

Read foundational work on documentation, evaluation, accountability and systemic risk.

Explore

Make risk accountable.

Connect observed events, plausible risks and control decisions without conflating them.

Explore

Build judgment before the consequential decision.

Draft learning guides for directors, committees and executives. These materials are educational and await editorial approval.

Explore

Make the method visible.

Versioned scope, evidence and review are prerequisites for defensible evaluation.

Explore

Comparable evidence comes before comparison.

No Responsible AI benchmark population has been approved for publication here.

Explore

Evidence supports a conclusion only within its stated scope.