Singapore’s Agentic AI Governance Framework

Singapore’s Model AI Governance Framework for Agentic AI provides guidance for organizations deploying systems that can plan, reason, use external tools, and take actions with limited direct supervision.

It emphasizes setting autonomy limits in light of a system’s context and risks, while preserving human accountability through approvals, oversight, and the ability to intervene. Its safeguards include pre-deployment and tool-interaction testing, ongoing monitoring, access controls, and logging. The guidance also addresses responsibilities across providers and users, and the added coordination challenges posed by multi-agent and third-party systems.