en.Wedoany.com Reported - The Postal Service Headquarters under the Ministry of Science and ICT announced on the 6th that it will upgrade the Postal Insurance AI Anti-Fraud System (IFDS) by introducing machine learning-based prediction models to identify increasingly sophisticated and organized insurance fraud. This project aims to shift from the existing detection method relying on fixed business rules to building an integrated prediction model that combines rule-based detection with AI technology, enabling the system to learn from diverse data and identify early signs of fraud anomalies.

As insurance fraud methods become increasingly organized and sophisticated, the scale of detected cases continues to grow. Losses caused by fraud can worsen insurance finances and drive up premiums, ultimately shifting the burden to regular policyholders, making it necessary to establish a more refined response system. The Postal Service Headquarters plans to adapt the operational experience and advanced application cases accumulated by private insurance companies in building anti-fraud systems to the business environment of Postal Insurance, thereby enhancing system effectiveness.

The project officially advanced following a kickoff meeting on June 12 and is the first initiative under the AI Transformation (AX) project promoted by Postal Insurance. Park In-hwan, head of the Postal Service Headquarters, stated that this project is not merely a system improvement but a starting point for AI and data-driven digital transformation, and that AI technology will be expanded to all areas of insurance business in the future to drive customer service innovation and operational efficiency. The Postal Service Headquarters expects that this move will enhance the prediction and detection capabilities for insurance fraud, while improving the efficiency of investigation and review processes, helping to prevent insurance fund leakage, protect policyholders, and enhance the trustworthiness and financial soundness of Postal Insurance.