Namespace: x-ai-sg:
Singapore AI Governance Framework
Jurisdiction: Singapore (SG) Frameworks addressed: Model AI Governance Framework (2nd Edition, 2020); Model AI Governance Framework for Generative AI (2024); Model AI Governance Framework for Agentic AI (January 2026); AI Verify testing framework (2022/2023); PDPC Advisory Guidelines on Use of Personal Data in AI Recommendation and Decision Systems (March 2024); MAS/ABS financial sector AI guidance. All frameworks are voluntary unless adopted by sector regulators. Registration tier: Open Review status: ai_reviewed Review date: 2026-03-20
Overview
Singapore's AI governance approach is voluntary and guidance-led. The frameworks are not binding horizontal legislation. Singapore launched the Model AI Governance Framework for Agentic AI in January 2026, supplementing the 2020 framework and the 2024 Generative AI framework. The PDPC issued updated Advisory Guidelines on AI and personal data in March 2024. MAS financial sector AI guidance has been updated through 2024 model risk management and GenAI guardrail work.
Fields
| Field | Type | Required | Description |
|---|---|---|---|
x-ai-sg:model_framework_version | enum | OPTIONAL | The Singapore Model AI Governance Framework version applied. Permitted values: model_ai_framework_2020, gen_ai_framework_2024, agentic_ai_framework_2026. |
x-ai-sg:model_ai_framework_applied_flag | boolean | OPTIONAL | Set to true where the AI system has been assessed against the applicable Singapore Model AI Governance Framework. |
x-ai-sg:ai_verify_tested_flag | boolean | OPTIONAL | Set to true where the AI system has been tested using the AI Verify testing toolkit against its eleven AI ethics principles and a test report generated. |
x-ai-sg:ai_verify_report_ref | string (URI) | CONDITIONAL | Required when ai_verify_tested_flag is true. Reference to the AI Verify test report. |
x-ai-sg:pdpc_ai_guidance_version_ref | string (URI) | OPTIONAL | Reference to the PDPC AI guidance version applied, including the Advisory Guidelines on Use of Personal Data in AI Recommendation and Decision Systems (March 2024). |
x-ai-sg:pdpc_advisory_applied_flag | boolean | OPTIONAL | Set to true where the AI system has been assessed against current PDPC advisory guidelines on AI and personal data protection. |
x-ai-sg:feat_principles_applied_flag | boolean | OPTIONAL | Set to true where the AI system is used in a financial services context and has been assessed against MAS FEAT principles and current MAS/ABS AI guidance. |
x-ai-sg:financial_sector_ai_guidance_ref | string (URI) | OPTIONAL | Reference to the specific MAS/ABS AI guidance applied, including model risk management and GenAI guardrail materials. |
Revision notes (from ai_reviewed pass, 2026-03-20)
model_framework_versionadded covering 2020, 2024 GenAI, and 2026 Agentic AI frameworks.ai_verify_report_refretained and made conditional onai_verify_tested_flag.pdpc_advisory_applied_flagdescription updated to reference March 2024 Advisory Guidelines specifically.pdpc_ai_guidance_version_refadded for version-specific reference.financial_sector_ai_guidance_refadded referencing updated MAS/ABS materials.- All frameworks described as voluntary throughout.
Citations
- PDPC Model AI Governance Framework (2nd Edition, 2020)
- IMDA/PDPC Model AI Governance Framework for Generative AI (2024)
- IMDA Singapore launches Model AI Governance Framework for Agentic AI, January 2026
- PDPC Advisory Guidelines on Use of Personal Data in AI Recommendation and Decision Systems, March 2024
- MAS information paper on AI model risk management (2024)
This namespace is registered and maintained by OPSF (Open Privacy Standards Foundation). Contact registry@pct.opsf.org for queries.