Fraud and risk management experts, The ai Corporation (ai), are helping global retailers to combat fraudulent use of fuel cards, and protect users, the company announced this week.

ai says its use of artificial intelligence (A.I.) and machine-learning (ML) could revolutionise the global retail industry’s use of fuel cards.

“The ai Corporation has developed a suite of self-service machine-learning products to enable fuel card issuers to reduce the impact of a number of fraudulent activities – including card copying, card swapping and skimming; at the same time as identifying and stopping fraud before it takes place,” the company said.

ai’s products are being used to help fuel card issuers prevent fraudulent activity in over 40 countries around the world, covering 150,000 sites and over 90 card and product types, and its unique A.I. and ML technology has already saved businesses a huge amount of money, ai said.

The company describes the technology as giving fuel card retailers the opportunity to write their own rule sets into any fraud platform, including ai’s own rules engine, RiskNet®. By using machine-learning, retailers can automate their fraud prevention, mitigating fuel card and transaction fraud, regardless of how and where it is perpetrated, ai said.

Matthew Attwell, Risk and Client Service Director at The ai Corporation, says: “ai’s blend of fraud expertise and innovative machine learning technology has been proven to deliver stand-out benefits to the fuel card sector.

“In a sector that involves sophisticated issuing relationships, with high multiples of cards for a single customer account, machine learning has proven it can deliver the high degree of accuracy to match, and then beat, the fraudster.”

According to ai, RiskNet can identify any type of fraud by creating specific rule sets for each type of fraud, whether it’s copied cards, card application fraud, driver-side collusion, the use of counterfeit cards or lost and stolen cards.

SmartScore, its says, is one of of six products within ai’s SmartSuite. It creates neural models using artificial intelligence and automated machine-learning techniques, to recognise patterns and trends in fuel card fraud, providing transaction risk scores to be used in conjunction with user-defined rules and parameter-mapping.

ai’s Link Analysis product can visually present the common links in historical transaction data, making the analysis of fuel card fraud much easier than traditional spreadsheet based approaches, the company says.