Feedzai has unveiled RiskFM (Risk Foundation Model), the industry’s first Tabular Foundation Model purpose-built for financial data and risk decisioning.
Banks have spent decades relying on rules and machine learning models built one customer at a time. RiskFM breaks that pattern: a single frontier model covering fraud detection, anti-money laundering (AML) and wider risk decisions across the full financial crime lifecycle. Unlike rival attempts limited to card network data, it is trained on a global dataset spanning onboarding, digital activity, payments, transfers and AML workflows.
Transactional data has resisted the foundation model wave for a reason. Language, audio and video are predictable enough for large language models to crack. Financial transactions are not.
“Next transactions are far less predictable than the next word in a sentence,” said Pedro Bizarro, chief science officer at Feedzai. “Consumer spending habits, payment types, and fraud modes change continuously. More importantly, financial risk is an adversarial domain; fraudsters actively adapt to evade detection in real time.”
Feedzai risk-assesses $9 trillion in payments across 120 billion events a year, from onboarding to real-time transfers. That breadth means RiskFM is tested at scale as one holistic model, not siloed in a single application.
Foundation model: What It Means
Early results stand out. RiskFM matches bespoke supervised models using data from a single customer, with no manual feature engineering, and beats them when trained across several institutions and geographies. It keeps improving as it ingests more data. For banks, that means faster deployment and lower implementation and maintenance costs.
“Foundation models have reshaped language, vision, and audio, but financial crime has remained stubbornly resistant to that wave,” said Sam Abadir, research director, risk, financial crime, and compliance for IDC. “Feedzai’s RiskFM is a credible attempt to close that gap. The early performance data is worth watching, as is how the model holds up as it expands into more complex use cases.”
“We’ve developed a foundation model for financial data that covers multiple use cases, from cards to real-time payments, and geographies, delivering strong performance from Day One at global scale,” said Pedro Barata, chief product officer at Feedzai. “We’re not just part of the conversation; we’re defining how it applies to the complexities of global financial crime prevention.”
Feedzai is validating RiskFM with early adopters, including Lloyds Banking Group. “We’ve been collaborating with Feedzai for years on AI innovation to give fraud fighters the upper hand against criminals, and RiskFM is an exciting milestone in that journey,” said Tom Martin, Lloyds Banking Group Business Platform Lead, Economic Crime Prevention.
Feedzai safeguards more than one billion consumers and $9 trillion in payment volume annually for the world’s top banks, payment networks and acquirers.
Related on Digital Trade Outlook
- Feedzai and Matrix USA partner to fight AI-enabled financial crime
- Razorpay Launches Vulcan, India’s First AI Payments Foundation Model, Built With NVIDIA and AWS
- Feedzai opens its $9 trillion fraud intelligence network to banks of all sizes