Tech & AI Awards Winner: Feedzai's AI-Powered Financial Crime Prevention
October 11, 2025 * 4 min read
Feedzai won the AI in Fintech Award at the Global FinTech Awards 2025 for its RiskOps platform, a single system used by banks and payment providers to detect fraud and money laundering. The company reports its platform protects more than US$70bn in annualised payment volume, scanning transactions to spot criminal activity in real time.
Why RiskOps matters for finance leaders
Risk often gets fragmented across fraud, AML, and authentication tools. Feedzai's RiskOps model brings these workflows together so teams can work from one source of truth and reduce vendor sprawl. That means fewer blind spots, faster case handling, and cleaner data for decisioning.
Digital payments generate a flood of signals. They tell a story about identity and intent, but they also give criminals room to hide. As Richard Harris, Head of Advisory at Feedzai, put it: "The key to success in digitised financial services is identity. Using the overwhelming amount of data to confirm identity is the single biggest challenge in combatting financial crime."
He added: "The core question we look at in Fincrime is 'are you who you say you are, and do you have permission to do the thing that you want to do?' That's fundamentally the challenge that we're solving for the industry, as identity is more and more complex and fragmented."
Responsible AI: performance and accountability
AI in financial services carries clear responsibilities. Decisions affect credit access, transaction approvals, and investigations, so fairness, transparency, and accountability must be built in from day one.
Pedro Bizarro, Chief Science Officer at Feedzai, summed it up: "You have fairness, you have accountability, transparency, you have the explainability part. And in addition to that: it still has to be performant and it has to be cost effective to put the solution in place." He also highlighted error distribution: "When we talk about fairness it's how we distribute errors in a way that we are not damaging particular population groups."
- Set clear metrics: fraud loss rate, false positive rate, customer friction, and segment-level fairness.
- Adopt explainability so analysts can justify outcomes to customers and regulators.
- Monitor error distribution across protected groups and channels; act on drift fast.
- Keep a human-in-the-loop for high-impact decisions; document overrides.
- Control cost-to-serve by tuning models, rules, and case queues together.
Digital euro: centralized fraud scoring at scale
Feedzai was selected as the first-ranked tenderer in a framework agreement with the European Central Bank to support a central fraud detection and prevention mechanism for the future digital euro. The approach provides a fraud risk score to payment service providers to complement their existing controls.
Nuno Sebastião, CEO and Co-Founder of Feedzai, said: "Being selected as the first-ranked tenderer in the framework agreement to secure the digital euro is both an honor and a responsibility. With tens of billions of transactions expected across the eurozone, success depends on AI that can adapt as quickly as fraud evolves."
He added: "Fraud isn't just numbers on a balance sheet. It's families losing their life savings and businesses losing customers. Protecting people and organisations from financial crime is why we built Feedzai. It's our steadfast mission to keep commerce safe."
For background on the initiative, see the ECB's overview of the digital euro project: European Central Bank: Digital euro.
What this means for banks, issuers, and PSPs
- Consolidate fraud and AML operations where possible to improve signal quality and speed of response.
- Prioritize identity truth: device, behavioral, payments, and KYC data should feed a shared risk graph.
- Track scams and authorized push payment fraud with models that learn from case outcomes and claims.
- Measure fairness and customer impact alongside loss metrics; treat these as first-class KPIs.
- Prepare model governance that meets regulator expectations across EU and US markets.
Quick facts
- Award: AI in Fintech, Global FinTech Awards 2025
- Platform focus: unified RiskOps for fraud and financial crime
- Scale: protects more than US$70bn in annualised payment volume
Next steps
If you're building your AI roadmap for fraud and AML, benchmark your tooling against unified RiskOps capabilities and the controls above. For practical resources on AI tools used in finance, explore this curated list: AI Tools for Finance.
Enjoy Ad-Free Experience
Your membership also unlocks:
