What Is a Deepfake? UK Fraud Losses Hit £1.28B as Banks Warn AI Makes Scams Harder to Spot

UK payment fraud losses hit £1.28 billion in 2025 as criminals increasingly use AI-powered deepfakes, cloned voices and synthetic identities to impersonate trusted people, according to UK Finance. Authorised push payment fraud rose 19% to £576.4 million, with banks warning that generative AI makes scams harder to detect.

What Is a Deepfake? UK Fraud Losses Hit £1.28B as Banks Warn AI Makes Scams Harder to Spot Banks say criminals are increasingly using deepfakes and synthetic identities to impersonate trusted people as authorised payment fraud rose 19% in 2025 UK payment fraud losses reached £1.28 billion $1.74 billion in 2025 as criminals increasingly turned to artificial intelligence to create more convincing scams, prompting fresh warnings from UK Finance, the trade association representing more than 300 banks, payment providers and financial services firms. The industry's Fraud Report 2026 showed authorised push payment APP fraud losses rose 19 per cent to £576.4 million $785 million last year, while overall payment fraud increased 4 per cent from 2024. UK Finance said organised criminal groups are increasingly using AI-powered tools, including deepfakes, cloned voices and synthetic identities, to impersonate trusted people and bypass identity checks. Once largely associated with manipulated celebrity videos and online misinformation, deepfakes are increasingly being used in fraud attempts targeting businesses and consumers, making scams harder to recognise before money is transferred. Why Banks Are Suddenly Worried About Deepfakes A deepfake is AI-generated video, audio or imagery designed to imitate a real person convincingly. Fraudsters increasingly use the technology to mimic company executives, relatives or financial institutions in an effort to persuade victims to transfer money or reveal sensitive information. UK Finance said criminals are also deploying voice cloning https://www.ibtimes.co.uk/connecticut-woman-thwarts-ai-voice-scam-1804247 , synthetic identities and so-called 'selfie injections', digitally manipulated images or videos submitted during remote identity checks to defeat fraud controls and open accounts using false identities. 'Fraud remains the most prevalent crime in the UK and is becoming increasingly sophisticated and internationally coordinated,' Ruth Ray, UK Finance's managing director of economic crime, said. Ray said organised criminal groups were rapidly adopting AI to automate attacks and increase their credibility, allowing scams to be carried out at greater scale while making them harder for victims to distinguish from legitimate communications. When AI Pretends To Be Your Boss Earlier this year, an employee at the Hong Kong office of Arup, the UK engineering consultancy, transferred HK$200 million, equivalent to about £19 million $26 million , after fraudsters used AI-generated video to impersonate senior executives during a conference call, according to Hong Kong police. Investigators said the attackers recreated company officials using publicly available video and audio before persuading the employee to approve multiple transfers. The case has become one of the clearest examples of how deepfake technology can be used to facilitate high-value corporate fraud. UK Finance identified similar tactics in its report, saying advances in generative AI are allowing criminals to produce increasingly convincing impersonation attempts that are harder to identify using traditional fraud checks. How Consumers Are Being Drawn In Businesses are not the only targets. Purchase scams accounted for £118.1 million $161 million in authorised push payment APP fraud losses during 2025, the highest of any authorised fraud category, according to the report. Investment scams, romance fraud https://www.ibtimes.co.uk/ai-romance-scam-california-couple-deaths-1804815 and account takeover attacks also remained significant sources of consumer losses, with criminals increasingly incorporating AI-generated content into their operations. Experian, the global consumer credit reporting and data analytics company, said businesses are engaged in what it described as an 'AI arms race' as fraudsters adopt increasingly sophisticated tools to create fake identities, forged documents and realistic impersonation attempts. Experian said organisations are responding by strengthening identity verification and fraud detection systems, although rapid advances in generative AI continue to narrow the gap between genuine and fraudulent communications. Banks Call For A United Front On AI Fraud Despite the growing sophistication of fraud attempts, banks prevented £1.68 billion $2.29 billion in unauthorised fraud during 2025 through fraud detection systems and other security measures, the report said. The UK government's Fraud Strategy 2026–2029 also identified generative AI as an emerging challenge. It warned the technology is lowering the barriers for criminals to launch increasingly credible scams at scale. UK Finance said responding to AI-enabled fraud will require closer cooperation between banks, technology companies, telecommunications providers and government. It also argued financial institutions alone cannot keep pace as criminal groups continue adopting new AI tools. Criminals are no longer relying solely on poorly written phishing emails or suspicious phone calls. The report points to a shift in the nature of fraud, with malicious actors able to recreate familiar voices, convincing video and trusted identities. Because of this, verification rather than appearance alone is now one of the strongest remaining defences. © Copyright IBTimes 2025. All rights reserved.