ILC ARTICLE

Prevention better than cure for insurance fraud

12th October 2021

While most fraud detection systems examine historical data to identify crimes after they have been committed, a new technology is being used by some of the world’s leading insurers to identify fraudsters before they strike.

Speaking during a Disrupting Fraud session at ClaimsTech, an I Love Claims event held in London, Digilog CEO Lior Koskas explained how his company’s Voice Risk Analysis (VRA) technology is an accurate and efficient process for validating conversation-based transactions in real-time.

He said: ‘DIVA (Digital Intelligence Voice Analysis) specialises in monitoring algorithms in speech to look for the differences between deception and truthful statements. This allows the claims handler to be alerted of potential risks during FNOL, enabling them to then ask more searching questions if necessary.’

Underlining the value of this service, he pointed out some startling statistics: people are lied to between 10 and 200 times a day, and during a 10-minute conversation with a stranger most people will tell four lies.

He said: ‘Fraud and deceptions costs the global economy $5.17tn every year, and insurance claims make up a big part of that.’

However, he insisted the solution is about more than fraud detection, explaining that improved customer service is an automatic benefit. He said that by identifying genuine claimants immediately insurers are able to fast-track their claims, with one insurance customer reporting a reduction in the lifecycle of honest claims from 30 days to four.

To date 25 UK insurers have embedded the technology into their business, while expansion is also accelerating in Europe and North America.

I Love Claims would like to thank ClaimsTech headline sponsor SightCall, as well partners Google Cloud and Mastercard, gold sponsors CoreLogic, Digilog UK, Entegral, IMS (Insurance Mobility Solutions), Tractable and Verisk, and silver sponsors asTech, Carpe Data, eviid, GT Motive, Service Certainty, and Synergy.