Shift Technology unveiled the third volume of its report Shift Insurance Perspectives. Belonging to the very closed club of unicorns, the company created in 2014 in Paris uses artificial intelligence and advanced data science techniques to detect suspicious behavior during claims reports and to automate them. With this global vision, Shift Technology offers with this new report concrete examples of significant fraud detected by AI.
Shift Technology has established itself as a leading player in AssurTech. Using its technology, the company analyzed hundreds of millions of claims and received the 2020 Frost & Sullivan Award for “Its best practices in reporting management solutions designed for insurance”. Established in 25 countries, its clients are major insurance groups such as AXA, AG2R-La Mondiale, Macif, etc. In the new edition of its report Shift Insurance Perspectives, the company shared five significant fraud patterns and indicates: “These few examples clearly demonstrate that AI is a technology that insurers can rely on to detect insurance fraud and take the right decisions more calmly”.
Insurance fraud upon taking out a contract
One insured has taken out two insurance policies, each covering a passenger vehicle, 24 hours apart. A few months later, it added 11 luxury cars to the second contract, via the insurer’s digital self-care tools. A link between multiple contracts was established thanks to AI and thus made it possible to discover that the insured managed a rental application between individuals. It insured the vehicles on behalf of the lessors, which was not part of the guarantees of the contract. This subscription fraud was detected and all contracts were terminated.
Auto insurance fraud
An insured declared a collision with an animal that caused significant damage to the front of his vehicle. The AI was able to make the link between the incident and a certificate of destruction of a vehicle which indicated the same damage as that declared. The anti-fraud service therefore investigated to find out whether the vehicle was already damaged and whether or not the insured had attempted to file a false declaration in order to have pre-existing damage repaired.
Health insurance fraud
An organized gang reported numerous claims, averaging over $ 5,000 each, and demanded reimbursement for medical treatment and benefits believed to have been received by traveling patients. The AI has made it possible to identify the connections between the members of the network and to detect the falsification of supporting documents, whether they are hospitalization reports or medical bills, pharmacies or medical analyzes …
Cancellation insurance fraud
An increase in attempts to fraud cancellation clauses has been observed since the health crisis. Fraudsters have used false medical reports (including positive COVID tests), forged tickets … to obtain compensation for allegedly canceled flights. AI has made it possible to detect suspicious activity by identifying those involved, establishing connections between policyholders and identifying false documents.
Easier to determine each person’s responsibilities in the event of a claim
The AI made it possible, after analyzing the documents in a compensation file, to determine that the fire in a rental apartment was linked to the recent installation, by the owner, of a boiler, that had been the subject of a product recall. The insurer was able to recover from a third party the costs initially incurred to compensate the insured and exonerate him.
–
