Allianz UK has developed an artificial intelligence (AI) tool to combat fraudulent insurance claims.
Called Incognito, the tool was developed in partnership with specialists from Allianz UK’s operations.
Go deeper with GlobalData
The AI-powered tool flags possibly fraudulent claims, which are submitted to a fraud specialist for thorough analysis and investigation.
Allianz UK said that Incognito has saved the company £1.7m since its creation, and an additional £3.4m is being held in claim reserves while the investigations are still ongoing.
The insurer claims that application fraud savings have risen by 150% in comparison with year-to-date forecasts.
It attributed the surge in fraud cases to a number of factors including the cost-of-living crisis, a weakened economy and the economic downturn.
The company is witnessing claims fraud across property, casualty and motor insurance lines of business.
The insurer added that in a sector where fraudsters are becoming more and more skilled, and the amounts, prices and kinds of fraud committed are changing, Allianz’s fraud prevention approach is starting to rely heavily on tools such as Incognito.
Allianz Commercial head of counter-fraud James Burge said: “The development of ‘Incognito’ has ensured that we enhance the market-leading service that we provide to our customers. We have been able to settle claims quicker and identify fraud at the earliest opportunity.”
Allianz Commercial head of claims data Alex Sorrie said: “The success of Incognito is shown through its cost-saving approach, consistent fraud detection rate over time, automation capabilities and regulatory compliance.
“Our partnership-driven approach and use of technology has set a new blueprint for combatting fraud and protecting our genuine customers and their premiums.”
Recently, Miller, a UK-based specialist reinsurance intermediary, launched an AI tool called MillerMo.
Based on technologies developed by Microsoft and OpenAI, it is designed to develop software, create content, translate, proofread and analyse data sets.
