AI-enabled fraud is escalating across payments, and Payment Expert spotlights the five RegTech firms using the technology to stop criminals before funds move.
AI changed how the payments industry operates, automating compliance processes and making payment experiences faster and more personalised. It changed how criminals operate too.
Bad actors upgraded their own strategies in a matter of years, moving from phishing attempts and stolen credentials to AI-generated identities, voice cloning and deepfake scams. RegTech and financial crime firms now use AI to identify suspicious behaviour, uncover hidden patterns and stop fraudulent payments before funds reach their destination.
The scale is visible in the numbers: per Deloitte‘s Center for Financial Services, generative AI could push US fraud losses to $40bn by 2027, up from $12.3bn in 2023, a compound annual growth rate of 32%.
Below, Payment Expert spotlights five companies leading the fight against AI-driven fraud.
Feedzai
As seen with AI, innovations create obvious advantages but can also introduce new challenges. Real-time payments have sped up the time it takes to send and receive transactions, but the window an institution has to identify fraudulent activity also shortens.
Established in 2011, Feedzai is a financial crime technology company with offices across the US, UK and Portugal, serving retail and commercial banks, payment service providers and merchant acquirers.
Feedzai uses AI to analyse the context surrounding a payment, harnessing transaction, behavioural, device and other information to determine whether activity is suspicious.
Detection traditionally ran on fixed rules, where transactions would be flagged if they met predetermined criteria, such as exceeding a certain value, originating from an unusual location or occurring at an unexpected time. However, Feedzai uses machine learning models to compare a payment with how a customer normally behaves across different channels.
Feedzai expanded its AI capabilities in 2026 with the launch of RiskFM, a foundation model made specifically for financial crime and risk decisioning.
RiskFM is designed to learn patterns across areas including onboarding, account activity, payments and transfers rather than relying on separate models for individual fraud scenarios.
BioCatch
Founded in 2011 and headquartered in Israel, BioCatch provides behavioural biometrics technology to banks, financial institutions and payments companies, with its tools also used across areas including e-commerce and digital identity.
A growing problem institutions face is that some fraudulent payments don’t look fraudulent.
In Authorised Push Payment scams, victims are manipulated into transferring money themselves, meaning they may successfully pass authentication checks and use their usual device.
BioCatch combats this through behavioural biometrics. Its technology analyses signals including typing speed, touchscreen gestures, mouse movements and navigation patterns to assess whether a user is behaving normally.
Machine learning can then compare current activity against previous behaviour and identify signs that a customer may be under the influence of a fraudster or that someone else has taken control of an account.
This approach is becoming increasingly important as criminals lean more towards social engineering instead of stealing passwords.
The value of this technology was proven in August 2026 when Visa agreed to acquire BioCatch for $2.4bn in cash. Visa said the acquisition would strengthen its fraud and security capabilities around account takeovers, scams, money mules and application fraud.
BioCatch is expected to continue operating with its existing leadership and structure as part of Visa’s Value-Added Services business following completion of the acquisition.
SEON
While BioCatch concentrates on behavioural biometrics, SEON uses a wider range of digital signals to determine the risk surrounding a customer and transaction.
Founded by Tamas Kadar and Bence Jendruszak, SEON operates globally across the Americas, EMEA and Asia-Pacific, providing fraud prevention and anti-money laundering (AML) technology to sectors including fintech, payments, e-commerce and iGaming.
The company’s fraud prevention technology analyses information including devices, behavioural patterns, velocity data, IP addresses and digital identity characteristics.
SEON’s AI scoring can use hundreds of data points to estimate the likelihood of fraud and help businesses find out if activity should be approved, reviewed or declined. The approach can be valuable against stolen payment cards, account takeover and synthetic identity fraud.
The company has also introduced technology to identify networks connecting accounts, devices and transactions.
Sardine
Payment fraud isn’t confined to one payment method, with fintechs and payment providers needing to monitor cards, bank transfers, instant payments and payouts while simultaneously assessing the identity and behaviour of the customer behind each transaction.
Founded in 2020 and headquartered in San Francisco, Sardine provides fraud prevention and AML technology to banks, fintechs, payment companies and other digital businesses.
Its platform uses machine learning to score transactions using information including identity, device, behavioural and historical payment data.
Models can also learn from outcomes such as declines, chargebacks and returns as fraud patterns change, which is useful when fraudulent activity stretches across different parts of the customer journey.
A criminal, for instance, could open an account using a synthetic identity, access it through a suspicious device and eventually move funds through an instant bank transfer.
Analysing these events separately can limit the information available to fraud teams, whereas connecting activity across onboarding, account usage and payments, AI provide a clearer picture of the risk surrounding a transaction.
ThetaRay
Another major challenge for fraud teams is separating genuine threats from the huge number of alerts generated by transaction monitoring systems.
Founded in 2013 and headquartered in Israel, with a presence in markets including the US, UK, Spain and Singapore, ThetaRay provides financial crime compliance technology to banks, fintechs, payment service providers and other financial institutions.
ThetaRay uses what it calls Cognitive AI to identify unusual patterns and relationships across large volumes of financial activity.
Its transaction monitoring technology analyses behaviour dynamically rather than relying entirely on static rules and predefined fraud scenarios. The company highlighted this approach in 2026 through its work with payments and merchant acquiring company Shift4.
According to a ThetaRay case study, implementing its AI-powered transaction monitoring technology resulted in an approximately 86% reduction in false positives and a 70% increase in productive alerts at Shift4.
Reducing unnecessary alerts and declines can help fraud teams concentrate resources on suspicious payments and limit friction for customers.
If you are interested in featuring in Payment Expert’s Spotlight series, get in touch with the team today by emailing News Editor Louis Thompsett at [email protected], or Senior Media Sales Executive Annabel Selvadurai at [email protected].
