- AI-powered fraud and financial crime prevention company, Featurespace, announces its results from highly anticipated pilot with Pay.UK to identify additional fraud in Authorised Push Payments.
- Featurespace’s proof of concept detected £138.7m of fraud at a 5:1 false positive ratio.
- Utilising their Generative AI solution (TallierLTM™), Featurespace later increased their fraud detection rates to 56%, detecting £180m of fraud, whilst maintaining a 5:1 false positive ratio.
- A key part of the solution is that it significantly minimises processing of sensitive data.
London, UK, 30th May 2024: Leading AI-powered fraud and financial crime prevention company, Featurespace, has today announced the results of a major pilot with the independent, not-for-profit operator of the UK’s national retail payments system including Faster Payments, Bacs and Direct Debit, Pay.UK, to protect UK consumers from Authorised Push Payment (APP) fraud.
The overall pilot exceeded expectations with participants’ results showing an average uplift of more than £112m worth of additional* fraud detected in a year. Featurespace significantly outperformed this average, achieving an uplift of £138.7m (43% of the missed fraud from incumbent fraud systems) at a 5:1 false positive ratio (within the data received). The initiative trialled a new overlay service which will allow all UK banks, building societies, and payment service providers (PSPs) to analyse money flows, utilising Featurespace’s Adaptive Behavioral Analytics to proactively detect fraud and help prevent crime before it occurs, safely and securely.
Cases of Authorised Push Payment (APP) fraud jumped by 12% from 2022 to 2023, with total losses amounting to £459.7 million**. APP fraud can occur when someone is tricked (scammed) into sending money to a fraudster.
Featurespace analysed data from participating banks, building societies, and payment service providers (PSPs) under a UK-first data-sharing agreement enabling Pay.UK’s ‘consortium’ model within their ARIC™ platform.
Post completion of the pilot, Featurespace ran the data through their Generative AI innovation (TallierLTM™) and was able to increase fraud detection to 56% at a 5:1 ratio. It is clear using this new approach has positive long-term implications for fraud detection.
Featurespace’s analytic services are available today and have already been deployed in banks and networks around the world.
Kate Frankish, Chief Business Development Officer and Anti-Fraud Lead at Pay.UK, commented:
“The positive results from this pilot demonstrate the importance of innovation and cross-industry collaboration in developing effective solutions to stay ahead of fraudsters and protect people in the ever-changing payments landscape. In 2023, the UK saw 232,429 people falling victim to fraud. To reduce the scale of the crime that is happening we need a unified approach, and this future service will be a major step forward.”
Martina King, CEO at Featurespace, commented:
“Fraud is the single largest crime in the UK. It accounts for 40% of all crime and contributes to £2.3 billion in losses annually. But the UK is leading the charge to tackle this issue and the game-changing pilot with Pay.UK is one that the world has been watching. It shows the immense power of collaboration and technology, and the scale of positive change that is possible when the payments industry works together to tackle fraudulent activity.”
A key part of Featurespace’s solution is that it significantly minimises processing of sensitive data.
Data privacy and security has been a core focus for Pay.UK, which has spent almost a year working closely with banks, building societies, PSPs, suppliers, and specialist legal counsel to ensure that all data is processed securely and legitimately within a strict scope of use.
