Despite higher inflation, interest rates and cost of living, the South African insurance market has seen a recovery over the past two years, with gross premiums written in the local insurance market expected to exhibit an annual growth rate of 5.14 percent from 2024 to 2028. (CAGR 2024-2028), reaching a market volume of $100.30 billion by 2028. As cloud adoption and of advanced data analytics remain catalysts to drive innovation, faster and more accurate intelligent decisions, and thus the growth of modern insurers, the rise and increased sophistication of insurance fraud remains a threat major.
Significant steps have been taken to combat fraud in local insurance markets. In its 2022 data report, the Association for Savings and Investments of South Africa (ASISA) recorded that overall, the local insurance sector recorded a loss of 77 million of rands due to fraud and dishonesty, but that losses amounting to R1.1 billion were avoided.
Despite this, with the advent and accessibility of artificial intelligence (AI) in recent years, SAS believes that the world is entering a dark era of fraud. Fraudsters are already leveraging sophisticated AI and machine learning-based tools to spread financial crimes. For example, AI and deepfake technology are helping fraudsters hone their multibillion-dollar skills. Phishing messages are more refined. Imitation websites look incredibly legitimate. A scammer can clone a voice with a few seconds of audio using simple online tools.
Given the size of the insurance market in South Africa, insurers have an incentive to act.
“Based on global benchmarks, fraudulent claims represent 3 to 7 percent of the total number of claims and up to 20 to 30 percent of the amount of claims paid. This is why auto, health and life insurers must constantly modernize their fraud detection systems to cope with the increasing sophistication of financial crimes also fueled by growing poverty,” says Stepan Vanin, regional head of business advisory at insurance, META and APAC at SAS.
He adds: “Fortunately, regulation, competition and the increasing complexity of fraud are pushing insurers to rethink how they can leverage technologies such as graphical analytics, statistical anomaly detection, computer vision and text analytics to protect their business operations while maintaining a high level of service to decent people. clients.”
Adopting AI-based solutions that integrate AI and machine learning capabilities is becoming essential. For example, SAS solutions enable insurers to detect suspicious activity throughout the lifecycle of a claim using a range of technologies. Sophisticated machine learning algorithms combined with a library of over 200 industry-validated business scenarios help detect strange claim circumstances and suspicious behavior patterns.
Automatic analysis of mathematical graphs adds a plus here by identifying typical collusion patterns between claim participants. Finally, computer vision can detect reused or corrected photos submitted in other claims months ago. The technology can also prevent potential fraudsters from purchasing new policies as soon as they are created. It enables insurers to collect, manage and analyze intelligence from any source to improve operational efficiency and effectiveness in the ongoing war against fraud.
AI gives insurers access to tools that collect and assimilate data and create relationships between variables they might never have considered. AI also helps insurers price and manage risks much better than before.
“An AI-based solution can help increase the accuracy and speed of fraud detection by agents using an analytical approach, anomaly detection techniques and machine learning. Using such an integrated platform can also centralize and automate fraud detection logic into a single decision point while reducing investigation time and increasing investigation efficiency,” says Itumeleng Nomlomo, Senior Business Solutions Manager at SAS in South Africa.
Currently, many insurers rely on simple approaches to detect agent fraud – at the application and claims settlement stages. Very few fraud cases are detected due to the lack of an integrated process between automatic detection and case investigation. SAS helped one of its global insurance clients increase the accuracy and speed of agent fraud detection using an analytical approach.
This centralized and automated fraud detection logic in a single decision point. The result was an increase of up to 40 percent in detected and confirmed agent fraud cases. This saved the company millions of dollars in just one year by avoiding losses and reduced investigation time from weeks to hours.
“With SAS, businesses can use AI and machine learning techniques to identify the types of insurance transactions that are likely to be fraudulent. AI techniques, including adaptive machine learning and unsupervised intelligent agents, can predict fraudulent transactions in real time based on changes and inconsistencies in customer behavior patterns,” adds Vanin.
Detecting insurance fraud faster will allow organizations to stop it sooner. More importantly, by reducing false positives, they can improve the effectiveness of investigations.
“Adopting a hybrid analytical approach that uses multi-layered detection methods to detect fraud during the individual claim or new business transaction phases will significantly help reduce the financial impact of fraud. By reducing false positives through AI and machine learning, insurers can use analytics to ensure that alerts highlighted for triage are much more likely to be provable fraud,” explains Nomlomo.
