Banking fraud to reach 2.2 billion transactions by 2031, fuelled by growing AI-enabled attacks
A new study by global tech strategists Juniper Research revealed fraudulent transactions in digital banking and money transfer will surpass 2.2 billion globally by 2031; up from 773.7 million in 2025, representing significant growth of over 180%.
The report identified growing AI-enabled attacks as changing the economics of fraud. By lowering costs of producing personalised social engineering attacks at scale, AI is making bank customers increasingly viable fraud targets. This represents an accelerating shift away from the compromising of banking infrastructure towards targeting end-user customers and authenticated payment journeys via social engineering attacks.
“Coupled with a reduced cost to commit fraud, AI enables fraudsters to create and adapt attacks faster, while instant payments reduces the window that banks have to identify suspicious behaviour. Banks must therefore combine behavioural, identity and payment intelligence prior to authorisation, rather than relying on controls after a payment has been initiated,” noted report author Shane O’Sullivan.
Agentic AI Raising Stakes for Banking Fraud Prevention
The report highlights how generative AI enables new and more adaptive scam typologies, while emerging agentic AI allows fraudsters to coordinate multi-stage attacks and adjust tactics dynamically as victims or controls respond. This increases the potential for fraud to become more persistent, personalised, and difficult to identify.
Banks must therefore move beyond detecting suspicious transactions towards further understanding how risks are developing across the customer journey. The report recommends banks use AI to build a continuous view of customer risk; analysing how behavioural, identity and payment signals evolve together to enable intervention before fraudulent transactions are authorised.
An extract from Banking Fraud Prevention Market 2026-2031, is now available as a free download.
