07/15/2019 | News release | Distributed by Public on 07/15/2019 04:03
My first posts in this series looked at how fraud is delivering a hard body blow to insurers' businesses and the obstacles they face in fighting this scourge. This post highlights the top seven tools for fighting insurance fraud:
Fraudulent tendencies can be detected automatically by an AI system. This includes powerful algorithms that collect data from social networks and cross-reference with other systems in seconds.
Using artificial intelligence to pick out inconsistencies and unusual patterns has quickly become standard for insurance companies, whether they're looking for sophisticated rings of fraudsters rigging auto accidents or just individuals embellishing how much their damaged property was worth,' writes Steven Melendez in Insights.
Data on fraud patterns has always existed. Now, machine learning can help insurance companies identify these fraud patterns in an automated way by unearthing exceptions and alerting the insurance companies to potential fraud before it takes place. Insurance companies can also use machine learning to study and identify suspicious claims that need to be further investigated. The potential for savings in this area is immense.
Predictive modeling/analytics can determine the probability of claims and analyze claims outside of the standard statistical range (exceptions), including geographical data mapping.
Claims reports usually include multiple pages, which made it difficult for text analytics to easily detect fraud in the past. But big data analytics can review unstructured data, to proactively detect fraud and speed the payment of legitimate claims.
Chatbots/virtual agents are the front-end for AI's power. For example, the peer-to-peer insurer Lemonade's claim-bot, A.I. Jim, gained some notoriety when it reportedly broke a world record by handling and paying out a straightforward claim within three seconds, with no paperwork required.
But what about fraud?
Within seconds, A.I. Jim reviewed the claim, cross-referenced it with the policy, ran 18 anti-fraud algorithms, approved the claim, sent wiring instructions to the bank, updated the policyholder and closed the claim.
Social media data mining offers a wealth of data that can help prove that a claimant is being less than truthful. 'This is an increasingly common situation, where we are able to use social media as part of our toolkit to be able to get rid of these claims,' said Catherine Burt, the national head of counter fraud at law firm DAC Beachcroft, which frequently works on personal injury cases for insurers, to The Guardian newspaper.
An automated underwriting solution with insurance-industry best practices and immediate access to critical information, plus key alerts, will reduce fraud. Fifty-five (55) percent of underwriting fraud attempts are uncovered currently, surging from 27 percent in 2012. We can mostly thank innovative underwriting systems for the improvement.
The aforementioned tools are all extremely helpful. But just as a master craftsman needs to create a plan of when and how to deploy all of his tools in an integrated fashion, insurers who achieve full integration of their core, data and digital systems and functionalities will be most successful in fighting fraud. Pre-integration equips insurers with a holistic look at customers, making fraud more apparent, and automates fraud detection. It also enriches ecosystems via leading insuretech technology, so insurers can benefit from cutting-edge fraud fighting tech.
Sapiens' NEW eBook, Fraud Fighting Tools for Insurers, provides a comprehensive picture of insurance fraud and explains how we can help.