Feedzai launches RiskFM, an AI model for detecting financial crime
Feedzai has released RiskFM, an AI model designed to improve how banks detect and prevent financial crime. The company describes it as the industry's first "Tabular Foundation Model" built specifically for financial data and risk decisions.
Banks typically build separate fraud detection and compliance models for each institution, relying on fixed rules and manually developed systems. RiskFM aims to replace that fragmented approach with a single model that works across different financial institutions.
The model targets a core pain point in compliance operations: the cost and complexity of maintaining institution-specific systems. Finance teams often spend significant resources building and updating these models without clear visibility into their effectiveness.
What this means for compliance teams
For finance and compliance professionals, RiskFM represents a shift toward standardized AI approaches to financial crime detection. Rather than custom-built models, banks could deploy a foundation model trained on broader financial datasets.
This approach mirrors how foundation models work in other industries-a base model trained on large datasets that can be adapted for specific use cases. In financial services, that means faster deployment and potentially more consistent detection across institutions.
The move also reflects growing pressure on banks to reduce compliance costs. Regulators continue to tighten anti-money laundering and financial crime rules, forcing institutions to invest more in detection systems. A standardized model could lower that operational burden.
Related reading
Your membership also unlocks:
