Artificial Intelligence and the future of financial fraud detection

Fraud has long been a major issue for financial services institutions, especially those that sell high volume, quick- turn-around-time-products such as funeral benefits. As new competing funeral, hospital cash-back and other similar products have entered the financial sector, the challenges relating to fraud have also increased. Fortunately, artificial intelligence has enormous potential to reduce claim fraud, as automated fraud detection tools get smarter and machine learning becomes more powerful, the outlook should improve exponentially.

XTND has identified and proved a total sum of R82 143 982 in funeral fraud over the last twelve months. One of the most prevalent and preventable forms of fraud is in the funeral insurance sector, which has been exacerbated by the increase in new products available in general retail stores. Buying a funeral policy has become easier than ever before, and in some cases, a financial advisor does not even form part of the policy lifecycle. The speed at which financial losses can occur when funeral claim fraud takes place, makes intelligent fraud detection techniques increasingly important.

Because of the availability of large volumes of customer data, together with proven fraud claim data that is updated in real time, AI can be used to effectively identify funeral fraud behaviour patterns that are irregular, for specific customers. XTND created an automated, predictive algorithm that specialises in determining the propensity of fraud on any funeral claim. An added advantage of a more sophisticated model like XTND’s VERITAS Propensity of Fraud Model, is its ability to use a wide variety of industry data to continually fit different customer claim thinking into the best-suited risk clusters, for more accurate identification of claims that need to be to be investigated. Thus, as the modus operandi and habits of claimants change, the model automatically adjusts what it views as potentially fraudulent claims. This identifies actual fraudulent claim transactions, and minimises false fraud flags (false positives).

False positives occur regularly with traditional rule-based anti-fraud measures, where the system flags anything that falls outside a given set of parameters. For example: if a claimant initiates a claim on a Monday, or if the deceased passed away at home, or if the claimant only provides a P.O. Box address as a residential address, this may trigger a fraud warning. A smarter system, such as XTND’s VERITAS model, (as described in the previous paragraph), better understands the underlying patterns of human behaviour, and can match a single claim against a different risk cluster of other insurers. In essence, XTND’s Veritas model ensures that the single claim’s behaviour is tested against multiple claims transactions before automatically raising a fraud flag.

By using VERITAS, insurers increase customer satisfaction as they are able to almost immediately pay low-risk funeral claims, preventing unnecessary interactions with honest claimants, limiting false fraud flags and ultimately reducing operational overheads.

The potential for funeral fraud will become even greater with the increase of new products and the eternal competition to give potential clients easy access to funeral cover. In addition, the newfound ability to abuse the emotional impact of the COVID-19 pandemic, makes it clear that it is imperative to use the most advanced techniques available, to fight funeral fraud and fraud on similar products.

Most exciting and motivating, is that we are now seeing that so-called opportunistic fraudsters are actually hard-core fraudsters through algorithms identifying them in previous fraudulent behaviour. This AI makes algorithms that were already intelligent, infinitely smarter.

Implementing the VERITAS Propensity of Fraud Model to curb funeral claim fraud, for example, will reduce insurer’s economic losses drastically. Fraud has been taking place throughout human history, and has only become more complex and difficult to stop as technology has advanced. Fortunately, we are now in a position where we are also able to leverage technology to identify fraudulent activities, and stop fraudsters before they cause harm. Achieving this will reduce insurers’ overall costs and improve their reputation with customers, who will likely be more loyal to an insurer that better protects their money and sustains a more long-term, affordable product premium. The possibility exists that insurers could channel some of the cost savings they make from reducing fraud, back to their low-risk customers in the form of lower premiums on funeral products. Ultimately, AI looks likely to create a radical shakeup of the entire financial industry, leading not only to reduced funeral fraud, but happier clients and greater customer advocacy from them. That truly is a win-win situation.