AI Verify Testing Framework and Toolkit
Is this legally binding?
Guidance. Voluntary guidance. Best practice, not obligation, until a contract or a regulator cites it.
Singapore's government-built AI governance testing framework (11 internationally accepted principles) plus open-source software toolkit for technical tests and process checks. Voluntary GUIDANCE tooling, self-assessment, not a legal requirement; launched as MVP by IMDA and PDPC on 25 May 2022.
From the source
“On 25 May 2022, AI Verify testing framework and software toolkit was launched as a Minimum Viable Product (MVP) for international pilot and feedback by IMDA and PDPC.”
What is AI Verify
Instrument record
- Short name
- AI Verify
- Date
- 2022-05-25
- Version
- Not versioned
- Cluster
- core
What this connects to
32 relations. Official relations are the ones the source documents state; anything marked GAGE analysis is our reading, not an agency's.
Operationalises1
- InstrumentModel AI Governance Framework, 2nd Edition (2020)GuidanceGAGE analysis, not official
Government-built testing tooling that operationalises Model AI Governance Framework principles into testable criteria
Operationalised by2
- ToolAI Verify Toolkit (Open Source)Guidance
The toolkit implements the testing framework's process checks and technical tests in software
- InstrumentProject Moonshot, LLM Evaluation ToolkitGuidanceGAGE analysis, not official
Moonshot extends the AI Verify testing approach from traditional AI to LLMs via benchmarking and red teaming
Issued by3
- AgencyInfocomm Media Development Authority (IMDA)Guidance
Launched by IMDA and PDPC, 25 May 2022
- AgencyPersonal Data Protection Commission (PDPC)Binding
Co-launched with PDPC, 25 May 2022
- AgencyAI Verify Foundation (AIVF)Guidance
Stewarded by AI Verify Foundation since June 2023
Cites4
- ToolAI Verify Toolkit (Open Source)Guidance
Structural decomposition of the source instrument
- ConceptTechnical Tests (Explainability, Robustness, Fairness)Guidance
Structural decomposition of the source instrument
- ConceptProcess Checks and Documentary EvidenceGuidance
Structural decomposition of the source instrument
- ConceptInternational Alignment of the Testing FrameworkGuidance
Structural decomposition of the source instrument
Cited by18
- SectionDimension 5: Testing and AssuranceGuidance
Section cites AI Verify Foundation and IMDA standardised model safety evaluations (Cataloguing LLM Evaluations)
- InstrumentPDPC Advisory Guidelines on Use of Personal Data in AI Recommendation and Decision Systems (1 Mar 2024)Guidance
Para 10.10 recommends AI Verify technical tests and process checks to support notifications and written policies
- ConceptUsing AI Verify to Support PDPA Compliance ClaimsGuidance
Para 10.10 recommends AI Verify to validate AI System performance
- Section12. Organisational ConsiderationsGuidance
Twelfth framework area in the AI Verify testing framework as used in official crosswalks
- EventAI Verify MVP Launched (25 May 2022)Guidance
- Section1. TransparencyGuidance
One of the 11 AI governance principles of the AI Verify testing framework
- Section2. ExplainabilityGuidance
One of the 11 AI governance principles of the AI Verify testing framework
- Section3. Repeatability / ReproducibilityGuidance
One of the 11 AI governance principles of the AI Verify testing framework
- Section4. SafetyGuidance
One of the 11 AI governance principles of the AI Verify testing framework
- Section5. SecurityGuidance
One of the 11 AI governance principles of the AI Verify testing framework
- Section6. RobustnessGuidance
One of the 11 AI governance principles of the AI Verify testing framework
- Section7. FairnessGuidance
One of the 11 AI governance principles of the AI Verify testing framework
- Section8. Data GovernanceGuidance
One of the 11 AI governance principles of the AI Verify testing framework
- Section9. AccountabilityGuidance
One of the 11 AI governance principles of the AI Verify testing framework
- Section10. Human Agency and OversightGuidance
One of the 11 AI governance principles of the AI Verify testing framework
- Section11. Inclusive Growth, Societal and Environmental Well-beingGuidance
One of the 11 AI governance principles of the AI Verify testing framework
- InstrumentConsultation Paper on Proposed Guidelines on Artificial Intelligence Risk Management for Financial Institutions (P017-2025)Consultation
CP para 2.7/footnote 10: Guidelines complement the initiatives under the AI Verify Foundation
- SectionAnnex ZA (national, informative): AI Verify as example testing toolStandard
National Annex ZA names AI Verify as an example voluntary testing tool
Maps across to3
- External frameworkNIST AI Risk Management Framework 1.0 (extern crosswalk target)Guidance
IMDA-NIST joint mapping exercise, crosswalk published Oct 2023 (AI RMF 1.0 functions to AI Verify 11 principles)
- External frameworkISO/IEC 42001:2023 - AI management system (extern crosswalk target)Standard
AI Verify Foundation crosswalk, Jun 2024 (ISO/IEC 42001:2023 Annex A controls to AI Verify testing framework)
- External frameworkFunction: MEASUREGuidanceGAGE analysis, not official
AI Verify's technical tests operationalise the MEASURE function
Mapped across from1
- SectionUse casesGuidanceGAGE analysis, not official
Project Moonshot (AI Verify Foundation) is a featured ASEAN use case
Verified against the official source on 2026-08-17. GAGE is not affiliated with or endorsed by any agency named here, and nothing on this page is legal advice. How this is built and checked.