News & Blog
AI helps in motor claims
detects fraud and saves millions on operational expence
Mains Lab and TK Service implement AI-powered quality control system for motor insurance claims

Moscow, March 30, 2019

Mains Lab and TK Service have implemented an IT platform powered by artificial intelligence, which helps insurance companies lower the number of mistakes they make during the appraisal of motor insurance claims by 30-40%.

TK Service is now capable of providing its partner insurance companies with 100% quality control for insurance claims calculations. The machine learning models can detect indirect links between facts describing a car accident and use them to detect fraudulent activities. Machine learning is also used to check photographs of a car after an accident and damages described in an insurance claim. If discrepancies are found, an expert is alerted to review the claim.

In the future, TK Expert plans to update the platform with features that allow conducting remote appraisals using only photographs of a damaged car. This is the first time that machine learning is used to appraise the cost of repairing a car in Russia.

Yury Kuvshinov, Chief Executive Officer of Mains Lab noted: "Machine learning is among the top 10 technologies that corporations use to increase the competitability of their business in the digital age. However, Russian insurance companies use no more than 15% of the available data in their decision-making process. This means that they're missing crucial insights that can be obtained from the data they already have. Using machine learning models helps insurers save millions of roubles by increasing effectiveness of business processes and levels of client satisfaction".

Ekaterina Miroshkina, Chief Executive Officer of TK Service said: "During the next two to three years the ability to accurately predict and appraise insurance claims will be the key factor in leading the market and keeping costs low. Our partner insurance companies aim to maximize automation of the insurance claim process, which results in higher technology standards for independent appraisers."