As the Monetary Authority of Singapore (MAS) recently published its Guidelines on Artificial Intelligence Risk Management for Financial Institutions, financial institutions are being advised to view this not simply as a compliance exercise, but as an opportunity to strengthen trust in AI adoption and build sustainable foundations for future innovation.
According to PwC, the guidelines signal that AI governance is entering a new phase, as while many organisations have established AI principles and pilot programmes, the challenge now is operationalising them at scale.
The MAS rules outlines the supervisory expectations to support the responsible adoption of AI, covering AI use cases, systems and models, including third-party AI and establish expectations across panning governance, risk assessment, AI lifecycle controls, third-party AI oversight, and organisational capabilities, reflecting both the growing importance of AI in financial services and the risks associated with its use.
Key takeaways:
- Strengthened governance and accountability – Clear board and senior management oversight to support responsible AI adoption
- Visibility into AI use – Robust AI inventories and risk assessments to help institutions understand where AI is used and the risks it may pose
- Lifecycle risk management – Proportionate controls across the AI lifecycle, from development and deployment to monitoring and oversight
- Accountability for third-party AI – Institutions remain responsible for risks arising from third-party AI models, services and providers
As part of this process, institutions are tasked to evaluate how AI may create, alter or amplify risks within their existing risk management frameworks, including:
- Financial risk: AI may lead to inaccurate decisions and financial losses
- Operational risk: AI may disrupt critical processes and controls
- Conduct risk: AI may create unfair or unsuitable customer outcomes
- Financial crime risk: AI may enable financial crime or reduce detection effectiveness
- Reputational risk: AI failures may damage trust and reputation
For enterprise-level CFOs across APAC, the guidelines highlight the growing need to balance AI-driven innovation with financial accountability, operational resilience and effective risk management.
As organisations expand their use of AI across financial planning, forecasting, reporting, fraud detection and decision-making, CFOs must ensure that adoption is supported by clear governance structures and measurable business outcomes.
Further, the guidelines also reinforce the importance of board and senior management oversight, comprehensive AI inventories and risk assessments, and controls throughout the AI lifecycle.
For finance leaders, these expectations provide a framework for evaluating whether AI investments can deliver sustainable value without introducing risks that undermine financial performance, regulatory compliance or stakeholder confidence.
The implications extend beyond financial institutions, particularly for enterprise CFOs overseeing AI-enabled finance functions and working with banks, technology providers and other third parties.
As AI becomes increasingly embedded in financial systems and outsourced services, organisations must understand how external models and platforms handle sensitive financial data, influence decisions and expose the business to operational or reputational risks.
Third-party AI adoption, in particular, raises questions about data security, accountability, model reliability and the ability to explain AI-generated outputs during audits.
CFOs can work with risk, compliance, technology and procurement teams to establish appropriate due diligence, monitoring mechanisms and accountability for AI-related decisions, ensuring that efficiency gains do not come at the expense of control and transparency.
Ultimately, the guidelines underscore a broader shift from AI experimentation towards disciplined implementation at scale.
