Fraudsters have never had it easier. It’s escalating and costing businesses billions annually. Tactics are becoming more sophisticated, and bad actors are armed with AI tools, stolen data and Dark Web networks.
A single fraudster can carry out refund fraud, place high-value fake orders, and teach others how to do the same – all in a day’s work.
Go deeper with GlobalData
As a result, UK retailers are facing unprecedented fraud losses. While many UK retailers are tightening their return and refund policies, more needs to be done to prevent fraud beyond these reactive measures. Tightening return and refund policies can have unintended consequences for loyal and legitimate customers, potentially damaging their experience and eroding trust
The hidden challenges retailers face every day
These fraud tactics aren’t happening in isolation. Dark Web forums act as an underground marketplace for stolen data, fraud advice, and even AI-driven scam tools to automate attacks. When a vulnerability is exposed, fraudsters don’t “gatekeep”it —they broadcast it to thousands of others who replicate the scheme at scale. Telegram, Reddit, and Dark Web forums are two popular platforms where fraudsters congregate, and even fraud DIY kits are being sold.
Today, almost anyone can become a fraudster with ease. Even a complete novice can quickly get the knowledge they need to rack up thousands of pounds from retailers in no time.
Trending fraud types on the rise
- Refund and returns fraud: Also known as ‘policy abuse’, whereby consumers knowingly exploit merchant terms for financial gain. This includes wardrobing (wear-and-return fraud), empty box scams, and false ‘item not received’ (INR) claims.
- Account Takeover Fraud (ATO): Stolen credentials are purchased from the Dark Web and used to place high-value orders, rack up debt, then disappear.
- AI-assisted fraud: Fraudsters using generative AI tools (like WormGPT) to automate attacks and get step-by-step advice for defrauding retailers.
How this is hurting UK retailers
Fraud is hitting retailers from all angles – financially, operationally and reputationally. While the direct financial losses are eye-watering, the British Retail Consortium estimates retail fraud cost businesses £1.04bn last year – the true costs extend far beyond chargebacks.
Retail is already a fast-paced and hypercompetitive industry with slim margins, meaning fraud-related losses hit harder. The pressure the current climate is having on fraud teams is immense too. Retailers relying on manual review processes can’t scale or respond fast enough to match fraudsters’ evolving tactics. In addition, manual processes are also highly prone to false declines, whereby legitimate transactions are mistakenly flagged as fraud, costing retailers legitimate sales and irritating legitimate customers.
The pain policy abuse is causing retailers is also evident in how many have tightened their return policies in recent months, with retailers reintroducing return fees, shortening grace periods, and tightening return acceptance criteria. ASOS’ crackdown on serial returners highlights the dilemma: stricter policies deter fraud but risk alienating good customers.
Zara and H&M now charge return fees, reflecting an industry-wide shift – but will these measures stop fraud or just shift it elsewhere?
Retailers are stuck in a lose-lose situation – too much fraud leads to heavy losses and stretched fraud teams, but excessive restrictions alienate customers. The challenge is finding a smarter way to differentiate bad actors from legitimate shoppers
AI’s role in fraud prevention
AI is transforming fraud prevention, allowing retailers to move from reactive strategies to a more proactive approach.
Machine-learning algorithms are replacing manual transaction reviews and can analyse behavioural patterns and flag anomalies in real-time. The most sophisticated platforms are also enabling retailers to leverage network-wide data across multiple retailers to detect fraud trends early and also identify the identities that are riskier, making more informed decisions in real time.
AI turns fraud prevention into a precision tool rather than a blunt instrument. Instead of blanket restrictions, AI assesses individual shopping behaviours – allowing retailers to apply targeted security measures and adjust policies dynamically.
In practice, that can look like:
- A long-time, high-value customer? Smooth, hassle-free returns.
- A suspicious shopper with multiple refund claims.
- Additional verification or return fees.
- A known abuser? Automatic rejection.
Winning the fight without losing loyal customers
Fraud is an ever-evolving and unavoidable reality in UK retail, but how retailers respond will determine their resilience and long-term success. Fraudsters will continue to evolve their methods – whether that’s the Dark Web, leveraging AI or finding new loopholes.
Retailers now need to shift from a defensive position to proactively managing risk. AI-driven fraud prevention is emerging as a way to strike this balance, identifying threats in real-time while safeguarding the seamless experiences for honest shoppers.
The fraud war isn’t slowing down – but neither is AI. Retailers that invest in intelligent, adaptive fraud prevention now won’t just minimise losses. They’ll gain the ultimate advantage: protecting their business while keeping customer experience at the heart of their strategy.
Aviram Ganor is General Manager EMEA at Riskified
