Singapore · framework

Singapore Model AI Governance Framework for Agentic AI

Governance framework focused on agentic AI, including bounding autonomy, human accountability, lifecycle controls and transparency.

voluntaryInfocomm Media Development AuthorityVerified 23 Aug 2026

Applicability

Organisations developing or deploying AI agents and agentic systems.

Enforcement context

Voluntary governance framework.

Regulatory timeline

Material dates and status changes.

20 May 2026

Agentic AI governance framework updated with deployment case studies

IMDA updated the framework with industry feedback, multi-agent and third-party-agent practices, automation-bias guidance and real-world case studies.

Source for this event →

22 Jan 2026

Singapore Agentic AI Framework

Governance framework focused on agentic AI, including bounding autonomy, human accountability, lifecycle controls and transparency.

Source for this event →

CRG implementation mapping

Operational obligations and evidence expectations.

SG-BOUND

Bound agent autonomy

Set explicit boundaries for what agents can access, decide and execute, including tool and data permissions.

Actor: developers/deployers · Domain: agent controls

Evidence: Agent permission model, allowlists, budget/time limits and escalation controls.

SG-HITL

Human approval checkpoints

Require human approval for high-impact, irreversible or materially consequential agent actions.

Actor: deployers · Domain: human oversight

Evidence: Approval matrix, intervention logs and exception records.

SG-LIFE

Agent lifecycle controls

Apply controlled development, testing, deployment, monitoring and retirement practices to agents.

Actor: developers/deployers · Domain: vendor lifecycle

Evidence: Lifecycle gates, versioning, whitelisting and rollback capability.

SG-TRANS

Agent transparency

Make agent roles, limitations and relevant automated actions understandable to users and accountable operators.

Actor: developers/deployers · Domain: transparency

Evidence: User disclosures, action traces and operator guidance.

Primary source

Verify the underlying instrument.

Cybatar's mapping is designed for AI governance and assurance work. Legal interpretation remains anchored to the current primary text and competent authority guidance.

Open IMDA Singapore source