AI fuels UK fraud surge as romance scams turn darker
TL;DR:
- More than four million UK fraud cases involving stolen money were reported last year — nearly eight a minute — with almost £1.3bn lost, according to UK Finance.
- Criminals are using AI to clone the voices of celebrities, family and friends, scaling deception and making victims more susceptible.
- UK Finance says banks cannot be “the only line of defence” and wants enforceable duties placed on tech platforms — a direct challenge for social media and marketplaces.
AI is making fraud cheaper, faster and harder to spot, and the toll is mounting. UK Finance’s annual report — the most comprehensive assessment of fraud losses in Britain — records more than 4.1 million cases where money was stolen last year, up 11% on the year and 31% on 2023, with almost £1.3bn taken. Investment scam losses alone jumped 40% to a record high.
Deception at scale
The report singles out AI as a force multiplier. Criminals are mimicking the voices of celebrities and even victims’ own relatives, lending fake approaches a credibility that lands at vulnerable moments. Romance scams have turned especially dark: UK Finance cites a case in which a fraudster married a victim to keep stealing, while a North Yorkshire florist lost £80,000 to a man using another person’s photos. Experts believe most scams go unreported, so the true figure is higher still.
The findings sharpen a debate Resultsense has tracked, following the Bank of England’s warning over AI deepfake scams that used manipulated video of its own governor. UK Finance’s Ruth Ray put the burden plainly: “The financial sector invests huge amounts in protecting customers, but we cannot be the only line of defence,” calling for “stronger, enforceable responsibilities” on platforms to remove fraudulent ads and verify sellers.
Looking forward
With experts predicting a wave of World Cup-related scams, pressure is building for tech platforms to act at source rather than leaving banks to refund losses after the fact — 12% of stolen money still goes unreimbursed. For UK consumers and businesses, the report is a reminder that AI-enabled fraud is now an everyday operational risk, not a future threat.