CAIA-FI-SG-F · Fundamentals grade · Assurance
Certified AI Assurance Financial Institutions Fundamentals — MAS
AI governed the MAS way: the Technology Risk Management Guidelines, the Guidelines on Artificial Intelligence Risk Management — including their agentic-AI scope — and the FEAT Principles, applied to AI systems in Singapore-regulated financial institutions.
2 hours examination · remote, proctored · valid 2 years
US$695
First membership year included
§ 1 — Intended candidates
Who this certification is for
The competency standard for AI assurance inside MAS-regulated financial institutions. Where a horizontal governance credential stops at principles, this one is specific to how a Singapore financial institution is actually expected to govern AI — the supervisory expectations a MAS-regulated firm answers to.
- Prerequisites
- No formal prerequisites. Candidates are expected to have professional experience in a risk, technology, audit, or compliance function.
This is the Fundamentals grade of the AI Assurance — Financial Institutions family — Professional and Fellow follow it. Also at Fundamentals in this family: Certified AI Assurance Financial Institutions Fundamentals — EU/DORA.
§ 2 — Examination domains
What the examination covers
The examination is mapped to MAS Technology Risk Management Guidelines, the MAS Guidelines on Artificial Intelligence Risk Management, and the FEAT Principles. Weightings are indicative until each syllabus is published.
- 01
MAS Technology Risk Management25%
The TRM framework applied to AI systems: technology governance, risk identification, and the controls a regulated institution is expected to evidence.
- 02
FEAT Principles and the Veritas methodology20%
Fairness, ethics, accountability, and transparency in financial AI, and the Veritas methodology for evidencing them on a live deployment.
- 03
MAS AI risk guidance20%
The MAS guidelines on AI risk management, including supervisory expectations for agentic AI in financial services.
- 04
Model risk management for AI/ML20%
Validation, monitoring, and documentation of AI and machine-learning models in a regulated financial institution.
- 05
Outsourcing, third parties, and accountability15%
Outsourcing, third-party and cloud dependencies, and board and senior-management accountability for AI.
Curriculum mapping indicates alignment of learning content with published frameworks. It does not constitute endorsement by, or certification under, any standards organisation or regulator.
§ 3 — Preparation
What preparation is anchored to
Candidates prepare independently against the published materials. The AI risk-guidance module is aligned to the final published guidance and revised on any material change.
The Council certifies individuals — it sells no preparation of any kind, and how a candidate prepared is never visible to an assessor. How impartiality is protected.
§ 4 — Assessment
How candidates are assessed
A two-hour scenario paper centred on a supplied deployment in a Singapore-regulated financial institution, producing a structured AI risk assessment against the MAS instruments the syllabus names.
Every script is marked twice — an AI assessor first and provisional, an independent human assessor second and final. The credential carries a unique identifier any employer can check on the public register.
§ 5 — Questions
Questions about this certification
Is this endorsed by MAS?
No. The curriculum is mapped to published MAS materials; mapping is not endorsement by, or certification under, any regulator.
Do I need to work at a bank?
No — the credential suits anyone assuring AI for MAS-regulated firms, including consultants and assurance providers.
What does the examination cost?
US$695, with the first year of membership included. Nothing is payable until you register for a sitting.
When can I sit the examination?
Registration is open now through the candidate portal; examination sittings run from the October 2026 window and are scheduled at booking.
Scheme-wide questions are answered on the FAQ and in the candidate handbook.
§ 6 — Further reading
Background reading
- How to build an AI risk profile for a deployment
A seven-step method for producing a risk profile a sceptical reader can challenge, with the failure mode at each step and what the finished artefact contains.
- AI assurance vs AI audit vs AI risk assessment
Three terms used interchangeably that mean different things, have different outputs, and require different independence. A practitioner-level distinction.
§ 7 — Register
Register for this examination
Create a candidate account and register now; sittings are scheduled from the October 2026 examination window and the examination fee is US$695, with the first year of membership included. Members pay a reduced rate on every examination.
Register — candidate portalThe Council is pursuing ISO/IEC 17024 accreditation; no accreditation is claimed until it is held. Accreditation status.