KT Gains Regulatory Approval for AI-Powered Voice Phishing Detection Service

Seoul, South Korea—KT Corp (NYSE: KT) announced on Friday that it has received regulatory approval for a real-time voice phishing detection service utilizing artificial intelligence. 

The service, approved by the Ministry of Science and ICT under the ICT Regulatory Sandbox program, uses on-device AI technology to detect voice phishing risk during live calls without requiring separate server connections. 

KT's new offering builds upon its existing 'voice phishing detection AI agent' by adding voice recognition technology. 

This enhancement compares suspicious calls with voice characteristics of past voice phishing criminals, analyzing voice similarity to alert users of potential risks. 

The AI language model detects financial fraud-related contexts in real time during calls.

Voice similarity measurement is conducted using data extracted from voice phishing samples provided by the National Forensic Service (NFS). 

The service operates entirely on-device to ensure user privacy, processing data without external server connections. 

The NFS collaborates with KT on this project, providing voice phishing data to improve detection performance. 

Following regulatory approval, KT plans to implement the technology in the spam-blocking app Who Who by VP Inc. 

This implementation will serve as a testing ground for the AI technology. 

KT aims to refine the AI technology and enhance the service through this real-world application. 

The company intends to use the insights gained from this deployment to improve further the accuracy and effectiveness of its voice phishing detection capabilities.