The Monetary Authority of Singapore (MAS) has released new AI Risk Management Guidelines, placing responsibility on bank board members and senior management to oversee risks arising from AI deployment.

The guidelines outline supervisory expectations for governance, key AI risk management systems, policies and procedures, lifecycle controls, and the capabilities and capacity needed to safely implement AI technologies, including Generative AI and AI agents.

MAS emphasizes that board members and senior executives must establish and enforce frameworks, policies, and processes for AI risk oversight, foster a strong risk culture, maintain accurate AI inventories, and conduct risk materiality assessments that consider impact, complexity, and reliance. Banks are expected to implement robust controls covering data management, fairness, transparency, explainability, human oversight, third-party risk, evaluation and testing, monitoring, and change management throughout the AI lifecycle.

Ho Hern Shin, MAS Deputy Managing Director, stated: “The proposed Guidelines on AI Risk Management provide financial institutions with clear supervisory expectations to support responsible use of AI in operations. These proportionate, risk-based guidelines enable innovation while ensuring key AI-related risks are addressed.” MAS is inviting feedback on the proposals from interested parties by 31 January 2026.