BioCatch has published UK survey findings showing banks face a growing challenge from AI-driven fraud. Most UK respondents said artificial intelligence is making scams harder to detect.
The UK results were based on responses from 80 fraud management, financial crime prevention, risk, and compliance professionals, and formed part of a wider survey of 1,440 banking leaders across 25 countries. Among UK respondents, 84% said AI had increased the sophistication of fraud and scam schemes, while 75% said it would be very challenging to distinguish legitimate AI-assisted banking activity from malicious or manipulated activity.
The figures suggest a shift in how banks assess suspicious behaviour. Rather than focusing solely on whether a customer has logged in with the correct credentials, institutions are increasingly being pressured to assess whether a payment reflects genuine intent or manipulation before money leaves an account.
Pressure on banks also appears to be rising. In the UK sample, 73% said fraud attempts at their institution were increasing, and 64% said fraud losses were rising.
Another finding focused on the receiving account in a payment chain. Some 76% of UK respondents said real-time intelligence on receiving accounts would significantly improve their ability to identify and stop scams.
AI pressure
The broader international results suggest concern about AI-led fraud is not confined to the UK. Globally, 88% of respondents said AI had already increased the sophistication of fraud, while 84% viewed AI agents as the industry's greatest exploitable vulnerability over the next year.
BioCatch said 80% of respondents reported that their institution had already encountered attacks using agentic AI. In addition, 60% of global respondents said AI-mediated banking would reduce the effectiveness of traditional fraud defences, while 72% said it would be very difficult to distinguish legitimate AI-assisted actions from malicious or manipulated AI activity in a future where AI agents commonly initiate transactions.
The UK response was slightly above the global average on one measure. Some 61% of UK respondents said widespread AI-mediated banking would reduce the effectiveness of traditional fraud signals.
The findings come as banks continue to deal with scams, mule accounts, and social engineering attacks while also handling reimbursement obligations tied to authorised push payment fraud. The data also suggest UK institutions may be seeing a different claims picture from peers elsewhere.
Only 19% of UK respondents said their organisation rejects more than USD $5 million in fraud claims annually, compared with 59% globally. Just 14% said their bank's customers lose more than USD $10 million to fraud and scams each year, compared with a global average of 39%.
This comes against the backdrop of UK reimbursement rules for authorised push payment fraud. According to figures cited in the research, 88% of APP scam losses, worth GBP £173 million, were reimbursed to victims in the first year of the policy.
Fraud trends
The global data also indicated that fraud pressure has intensified over the past year. In the previous edition of the survey, 71% of respondents reported an increase in fraud attempts at their organisation, rising to 81% in the latest research.
Reports of rising year-on-year fraud losses also climbed, from 59% to 76%. Nearly half of those surveyed said their organisation loses more than USD $10 million a year to fraud, 20% said losses exceed USD $25 million, and 5% said they exceed USD $50 million.
Concerns about speed featured prominently. Globally, 76% of respondents said they were very concerned about the increasing pace of fraud in their region, while 70% of UK respondents said the same.
The survey also touched on the customer impact of anti-fraud controls. More than 96% of respondents said their institution already measures customer attrition linked specifically to fraud and scam experiences, while 39% said it is a primary driver of investment decisions.
Globally, 68% of banking leaders said their organisation's approach to fraud prevention and reimbursement had resulted in a net loss of customers. Of those, 56% linked the loss to unreimbursed claims, while 44% blamed excessive friction.
Jonathan Frost, Global Advisory, EMEA, BioCatch, said the pressure on banks is moving earlier in the payment journey.
"Agentic AI is making fraud faster, more scalable, and harder to detect," Frost said. "UK banks should prioritise early prevention in the payment process. Although strong reimbursement rules protect victims after fraud, they do not prevent emotional harm, disrupt mule accounts, or stop criminals from profiting. Criminals will inevitably use AI, potentially leading to exponential growth in fraud. Only behavioural insights, shared in real time across the sector, can detect customer manipulation, assess user intent, identify mules, and support a risk-based approach to friction."
Gadi Mazor, Chief Executive Officer of BioCatch, said the changes go beyond fraud teams.
"AI is starting to reshape how customers interact with eCommerce sites and financial institutions and will change how criminals execute fraud and other financial crimes," Mazor said. "As digital interactions continue to grow faster, more automated, and increasingly driven by agents, we must move beyond static identity checks and toward a deeper and immediate understanding of behaviour, intent, and trust."