top of page

When AI Meets Finance: What MAS's New Risk Framework Means for You

A Report by CYS Global Remit FinTech Development Unit


Singapore's financial regulator has set clear expectations for how AI must be governed in financial institutions. If your business moves money, these rules are relevant to you.


A New Era of AI Oversight

Artificial intelligence is no longer an experiment on the fringes of finance. It now powers fraud detection, customer onboarding, FX pricing, and transaction monitoring across Singapore's financial sector. Recognising this shift, the Monetary Authority of Singapore (MAS) has moved decisively to ensure that the rapid adoption of AI is matched by equally robust governance.


In November 2025, MAS issued a consultation paper proposing Guidelines on Artificial Intelligence Risk Management — the first sector-wide AI governance framework for financial institutions in Singapore. The consultation closed in January 2026, and MAS is currently reviewing responses as it finalises the guidelines.


Project MindForge: Industry Working Together

Alongside the proposed guidelines, MAS concluded Phase Two of Project MindForge in March 2026. It is a collaborative initiative involving 24 leading financial institutions, insurers, capital market firms, and technology partners including DBS, OCBC, HSBC, AIA, BlackRock, and Microsoft


The result is a practical AI Risk Management Toolkit featuring an Operationalisation Handbook. It functions as a step-by-step guide for financial institutions to implement AI governance frameworks. The Handbook covers four key areas:


  • Scope and oversight: Who is responsible for AI governance at board and management level

  • Risk identification and assessment: How to evaluate the materiality and nature of AI risk 

  • AI lifecycle controls: Managing AI systems from development through deployment to retirement

  • Capabilities and capacity: Ensuring staff, infrastructure, and skills are fit for purpose


What the Guidelines Require

The proposed MAS guidelines apply to all financial institutions(Banks, Insurers, Capital Markets firms, and Payment Service Providers,etc…). They are designed to be proportionate, meaning smaller or less AI-dependent firms face lighter requirements than large institutions with AI embedded in core operations.


Key expectations include:


  • Board-level accountability for AI risk governance

  • A comprehensive and up-to-date AI inventory covering all systems in use

  • Structured risk assessments before deploying AI in customer-facing or operational processes

  • Ongoing monitoring and clear escalation procedures when AI systems underperform or behave unexpectedly

  • Governance obligations that extend to third-party AI tools - institutions cannot delegate responsibility to vendors


The guidelines explicitly address generative AI and agentic AI — autonomous systems that can take actions on behalf of the institution — reflecting how rapidly the technology landscape is evolving.

Key Principle from MAS: Institutions must maintain sufficient AI literacy at the board and senior management level. AI governance is not just a technical matter — it is a leadership responsibility.

What This Means for Cross-Border Payments

For businesses that use regulated payment service providers like us, these developments carry a clear signal: the institutions handling your money are being held to an increasingly rigorous standard of AI governance.


AI is already present in how payment providers manage compliance screening, detect unusual transactions, and route funds across corridors. MAS's framework ensures that these AI systems are not black boxes - that they are tested, monitored, governed, and explainable when questions arise.


For SMEs and high-net-worth individuals, this will eventually translate into greater confidence. This ensures that decisions that are made about your transactions (Flagged/Delayed/Approved Payments) are made within a structured and accountable framework. 


What Comes Next

MAS is expected to finalise the AI Risk Management Guidelines following the consultation review. A 12-month transition period is proposed, giving institutions time to align their practices before formal supervisory expectations apply.


MAS has also established a BuildFin.ai initiative to bring together financial institutions, technology providers, and research bodies to tackle shared AI governance challenges, including the emerging risks of agentic AI. Singapore is positioning itself as a global benchmark for responsible AI in finance - and that benefits everyone who relies on its financial system.


Sources

Monetary Authority of Singapore (MAS) — Consultation Paper on Guidelines on Artificial Intelligence Risk Management (November 2025): https://www.mas.gov.sg/publications/consultations/2025/consultation-paper-on-guidelines-on-artificial-intelligence-risk-management


MAS — Project MindForge AI Risk Management Toolkit (March 2026): https://www.mas.gov.sg/schemes-and-initiatives/project-mindforge


MAS — Media Release: MAS Partners Industry to Develop AI Risk Management Toolkit for the Financial Sector (20 March 2026): https://www.mas.gov.sg/news/media-releases/2026/mas-partners-industry-to-develop-ai-risk-management-toolkit-for-the-financial-sector

bottom of page