UK and Singapore team up on AI in finance, putting safety first and enabling cross-border growth

UK-Singapore AI-in-Finance Partnership: What It Means for Banks, Insurers, and CFOs

The Monetary Authority of Singapore (MAS) and the UK Financial Conduct Authority (FCA) have formalised a cross-border partnership to accelerate safe, responsible AI use in financial services. Announced at the Singapore FinTech Festival, the UK-Singapore AI-in-Finance Partnership gives firms in both markets a clearer path to test, deploy, and scale AI under aligned supervisory expectations.

What the partnership covers

The agreement opens the door for AI solution providers to work across both regulatory environments and for financial institutions to collaborate on pilots and share learnings. The goal: reduce friction, raise standards, and help firms move from isolated experiments to production-grade AI with proper guardrails.

Jessica Rusu, chief data, information, and intelligence officer at the FCA, said the partnership will help firms "grow through collaboration, gauge new cross-border opportunities, and shape the future of responsible AI innovation in finance." Kenneth Gay, chief fintech officer at MAS, noted the shift "from experiments to enterprise use, and from individual models to connected, agentic systems," with safety and scalability as priorities.

FCA to establish a Singapore presence

The FCA will place a Financial Services Attaché at the British High Commission in Singapore-its first formal foothold in the country. Expect tighter coordination with MAS, faster feedback loops for firms, and stronger support for cross-border pilots and market entry.

Joint workstreams and programs

Planned activities include joint testing of AI applications, supervisory information sharing, and industry events focused on AI in finance. The agencies will leverage MAS PathFin.ai and the FCA's AI Spotlight to exchange solutions and coordinate engagement with firms.

Learn more about the regulators here: Monetary Authority of Singapore and UK Financial Conduct Authority.

MAS consultation: AI risk management guidelines

In parallel, MAS released a consultation paper proposing guidelines that clarify supervisory expectations for AI risk management across the lifecycle. The framework is proportionate-requirements scale with the materiality and complexity of your AI use.

Key focus areas for boards, senior management, and risk leaders:

• Governance: Clear accountability for AI risk at the board and executive levels, with defined lines of ownership.
• AI inventory: A comprehensive, living register of AI use cases, models, data sources, and third-party components.
• Risk identification: Upfront assessments covering model, data, operational, conduct, and consumer risks.
• Controls: Policies and tooling for data quality, privacy, fairness, explainability, performance monitoring, and human oversight.
• Lifecycle management: Testing and validation pre-launch and ongoing monitoring post-launch, including change management.
• Third-party risk: Due diligence, contractual controls, and contingency plans for vendors and open-source components.

MAS is seeking feedback until Jan. 31, 2026. The guidelines draw from recent supervisory reviews and ongoing industry dialogue.

Why this matters for finance, insurance, and management teams

• Faster cross-border pilots: A clearer route to test and scale AI solutions across Singapore and the UK with aligned supervision.
• Lower compliance uncertainty: Early clarity on what "good" looks like for governance, fairness, and human-in-the-loop controls.
• Better vendor leverage: AI providers operating in both markets can align to common expectations, reducing integration overhead.
• Operational resilience: Stronger model and data controls reduce conduct, bias, and outage risk.

Action checklist for leaders

• Brief your board and risk committee on the partnership and the MAS consultation timelines.
• Stand up or refresh your AI inventory; flag high-impact use cases (credit, underwriting, claims, AML, fraud) for priority controls.
• Map current practices to the proposed MAS guidelines across governance, data, testing, monitoring, and human oversight.
• Set up joint workstreams with Compliance, Risk, Data, and IT to close gaps before large-scale rollout.
• Engage trusted vendors that can meet both MAS and FCA expectations; update contracts with control and audit clauses.
• Prepare for cross-border pilots via MAS PathFin.ai and FCA AI Spotlight; document testing protocols and outcomes.
• Submit feedback to MAS where requirements impact core processes, legacy systems, or third-party dependencies.

Compliance and build-readiness: practical moves

• Documentation: Keep model cards, data lineage, validation results, and change logs audit-ready.
• Fairness testing: Define metrics by use case; set thresholds and escalation paths for adverse results.
• Human oversight: Make intervention points explicit, especially for decisions affecting customers and capital.
• Monitoring: Track drift, performance, and complaints; automate alerts and periodic reviews.

If your teams need a structured way to upskill for regulated AI deployment, explore role-based learning paths and vetted tooling for finance: Courses by job and AI tools for finance.

Bottom line: the UK-Singapore partnership lowers the barrier to safe AI adoption across two major markets. Firms that align early on governance and lifecycle controls will move faster-with fewer surprises.