Revolut launches research division to build proprietary AI for fraud detection and financial services
Revolut has launched Revolut Research, a dedicated division within its AI Department focused on developing machine learning technologies for financial services, as the digital banking company moves to expand the use of artificial intelligence across its global operations.
The new research unit will work with academic and technology institutions to develop machine learning architectures and provide a common foundation for Revolut’s proprietary AI systems. The company says the initiative is intended to move AI development deeper into its financial infrastructure rather than relying primarily on third-party software layered onto existing banking systems.
At the centre of the programme is PRAGMA, Revolut’s proprietary foundation model developed in collaboration with NVIDIA. The model is designed to analyze complex financial behavior and support applications including real-time risk assessment, fraud detection, platform operations and personalised financial product recommendations.
Revolut said testing PRAGMA against historical data produced significant improvements compared with legacy models. The company reported a 2.3-fold increase in accuracy when identifying credit default risk, a 65% increase in fraud cases detected alongside a 17% improvement in alert precision, and a 41% improvement in the relevance of product recommendations across retail and business accounts.
The company attributes part of the model’s potential to the scale and diversity of its financial data. Revolut serves more than 80 million retail customers across over 40 markets and processes billions of cross-border transactions. According to the company, these transactions and associated financial behaviors provide data that can be used to train models for applications such as fraud prevention, credit risk assessment and customer recommendations.
Revolut Research is being led by Anton Repushko, who has presented the division’s research at international machine learning conferences, including the International Conference on Machine Learning (ICML) in Seoul.
The company is also expanding the use of AI beyond its foundation model research. Its security systems analyse nearly one billion transactions each month to identify potentially fraudulent activity, while its AI assistant, AIR, can perform multiple financial tasks within the Revolut app. AIR is currently available in the UK.
Revolut’s Head of AI, Pavel Nesterov, said “To lead the future of intelligent banking, you cannot rely on third-party blueprints. We have launched Revolut Research to institutionalise our ‘build, don’t bolt on’ philosophy. By training native foundation models on our global operational data, we are giving our engineering teams an unprecedented engine to deploy smarter features faster, eliminate systemic friction, and give our customers a safer, radically better financial experience.”
The company plans to publish research findings and open source selected technical frameworks as the division develops. Revolut Research is also expected to participate in scientific and technology events, including NVIDIA GTC Berlin in October and the ICAIF conference in November, while hosting quarterly meetings for members of the scientific community.
The launch positions Revolut among financial technology companies investing in proprietary AI infrastructure as banks and other financial institutions accelerate efforts to integrate machine learning into risk management, fraud prevention and customer-facing services.
Download the FREE Kaspersky Next Enterprise Security Guide here to explore the complete framework for simplifying security operations and building cyber resilience.
Go to TECHTRENDSKE.co.ke for more tech and business news from the African continent and across the world.
Follow us on WhatsApp, Telegram, Twitter, and Facebook, or subscribe to our weekly newsletter to ensure you don’t miss out on any future updates. Send tips to editorial@techtrendsmedia.co.ke
