MAS is moving from principles and industry toolkits towards detailed, proportionate AI risk-management expectations for all financial institutions, including GenAI and agents.
The consultation proposes board oversight, AI inventories, materiality assessments and lifecycle controls. MindForge provides practical implementation material while final guidance is pending.
The proposed guidelines apply across MAS-regulated financial institutions, proportionate to size, use and risk profile.
Summarised from the primary instruments below. Open the source before relying on an obligation for legal or compliance decisions.
Board and senior management oversight, clear ownership, risk culture and policies proportionate to AI adoption.
Inventory AI, assess materiality and apply testing, evaluation, monitoring and change controls across the lifecycle.
Apply the framework to GenAI and agents, with safeguards calibrated to autonomy, reliance and potential impact.
The original authority remains the source of truth. Dates below are publication and effective dates, not OneBench ingest dates.
Monetary Authority of Singapore
Proposes sector-wide expectations for oversight, inventories, materiality and lifecycle controls covering traditional, generative and agentic AI.
Monetary Authority of Singapore
Provides an operational handbook and case studies aligned to MAS’s proposed risk-management expectations.
Monetary Authority of Singapore
Industry work led by MAS to develop runtime safeguards for agentic AI in finance, complementing lifecycle governance and testing controls.