Feedzai introduces AI Model to transform financial risk detection

Feedzai has launched RiskFM (Risk Foundation Model), a new AI solution designed specifically for financial data and risk decision-making, marking a significant step forward in the detection and prevention of financial crime.

 

With a strong presence in Porto, the company continues to help financial institutions identify and prevent fraud, money laundering and other financial risks with greater speed, reduced manual effort and enhanced adaptability. Unlike traditional approaches based on manually developed rules or client-specific machine learning models, RiskFM introduces a new paradigm by leveraging a global dataset covering onboarding, digital activity, payments, transfers and anti-money laundering (AML) processes. This allows institutions to detect and respond to evolving financial threats with greater accuracy and consistency.

 

According to Pedro Bizarro, financial transactions present unique challenges compared to other AI domains, due to their unpredictability and the adaptive nature of fraud. RiskFM addresses this by continuously learning from diverse datasets, improving its performance over time.

 

The model has demonstrated strong results in testing, outperforming traditional machine learning approaches and delivering high performance without requiring extensive manual feature engineering. It can also be customised for individual institutions while maintaining scalability across multiple use cases, from fraud detection to AML compliance.

 

With this launch, Feedzai reinforces Porto’s position as a hub for fintech innovation, contributing to the development of more secure and resilient global financial systems.