MAS Partners Banks and Police to Deploy AI Against Financial Scams

Singapore's financial regulator is going on the offensive against scams. The Monetary Authority of Singapore (MAS) announced a collaboration with five banks, the Government Technology Agency, and the Singapore Police Force to deploy AI and machine learning for the pre-emptive detection of financial fraud.

At the core of the initiative is a Proof-of-Value (POV) exercise that aggregates historical transaction data from participating banks to train AI models capable of identifying high-risk transactions and accounts before damage is done. The early detection approach is designed to enable rapid intervention and reduce customer losses.

MAS has established a secure data-sharing environment with strict governance. Bank account numbers are hashed using one-way cryptographic techniques so that only the originating bank can identify actual accounts. Data access is restricted to authorised personnel within a continuously monitored environment, and all data will be deleted upon completion of the POV.

The initiative lays groundwork for deeper sector-wide collaboration, with MAS considering expanded scope and more sophisticated models incorporating broader datasets if the POV proves effective. It comes as part of a broader push to harness AI across the financial sector, following the conclusion of Project MindForge Phase 2 in March, which produced an AI Risk Management Toolkit developed with 24 financial institutions.

Why it matters for Singapore: Scams cost Singaporeans hundreds of millions of dollars annually, and this marks the first time the regulator has pooled bank data at scale for AI-driven fraud detection. If successful, it could become a template for other financial centres grappling with the same problem.