Predictive modeling to fight insurance fraud

In the United States, insurance fraud causes an annual loss of nearly 80 billion USD. To cope with this situation, insurers are increasingly using AI (artificial intelligence).

A study published by the Coalition Against Insurance Fraud shows that 80% of American insurers use predictive modeling to detect possible fraud. In 2018, this rate did not exceed 55%. Text mining (automatic language processing) is also rising in popularity with 65% of insurers using it in 2021 compared to 33% in 2018.

To achieve good results in the fight against insurance fraud, companies are diversifying their data sources and no longer limiting themselves to internally generated data.

Increased use of identity verification and photo analysis software is also among the trendy technological methods.