Samsung Electronics, 3 mobile carriers detect voice phishing in real time while on the phone with AI

*This content was translated by AI.

The Ministry of Science and ICT announced on the 12th that it would like the public to use the AI-based voice phishing detection notification service provided by Samsung Electronics and three mobile carriers to prevent voice phishing damage.

First of all, Samsung Electronics is providing a "voice phishing suspicious phone notification" function in a phone app built into a Galaxy smartphone. This function analyzes the call contents of a phone call made to an unknown number and provides notifications to users in two stages: "Doubt (Voice Phishing Suspicion)" and "Warning (Voice Phishing Detection)."

It is available on smartphones with Samsung Galaxy One UI 8.0 or higher. Samsung Electronics is providing the function by default so that many users can use it, and when users do not want it, they can deactivate the function in the app setting.

SK Telecom is providing a function that analyzes conversations in real time during a call through the "Aida Phone" app and informs whether voice phishing occurs. It comprehensively analyzes various factors such as whether suspicious keywords are included and conversation patterns, and divides the results into two stages, "doubt" and "danger," and informs users by warning pop-ups, notification sounds, and vibrations during the call.

The A-dot phone app provides a "real-time AI spam phishing detection" function that shows spam and phishing warning information when receiving calls from unknown numbers and "no/no" information evaluated by users within the app. ▲ A "dangerous phone call warning" function that displays a warning notification when sending to a number detected by spam phishing ▲ A "phone interception detection notification" function that identifies and informs as soon as a specific app intercepts a call.

KT is providing a voice phishing detection and notification service that combines real-time context detection, speaker recognition, and deep voice detection technology during calls through the 'Who Who' app. Context detection is a technology in which an artificial intelligence that has learned a voice phishing crime scenario analyzes a call voice to inform whether it is phishing, and speaker recognition and deep voice detection are technologies to find a reported voice phishing criminal's voice or falsified voice.

KT evaluated that the real-time voice phishing detection function had the effect of preventing 3,000 voice phishing out of a total of more than 46.8 million call traffic in 2025. In addition, it said that the detection accuracy increased from 90.3% in the first quarter of 2025 to 97.2% in the fourth quarter of 2025 by continuously upgrading the features.

LG U+ is providing a voice phishing detection notification service during the call through the "ixi-O" app. In addition to simply analyzing conversation patterns, the "Anti-Deep Voice" function, which identifies artificial intelligence falsified voices, and the "Criminal Voice Detection" function, which detects whether or not they match the reported criminal voice (sexual), work together to notify users with a warning pop-up and a notification sound if voice phishing is suspected.

The Exio app provides a "automatic spam phone AI reception" function in which artificial intelligence responds to calls made to ▲ unknown numbers instead of ▲ malicious web addresses (URLs) in KakaoTalk and a "dangerous URL and malicious app detection" function that detects malicious app installations. LG U+ also plans to introduce a function that guides suspected voice phishing numbers in advance so that customers can block risky calls before making a call.

Choi Woo-hyuk, head of the network policy division at the Ministry of Science and ICT, said, "Recently, voice phishing is characterized by sophisticated techniques using artificial intelligence and maximizing psychological pressure. In order to prevent damage, it is important to share these latest techniques and coping experiences with acquaintances and to be alert," adding, "Since various types of voice phishing crimes such as impersonation of parcel delivery, impersonation of family members, and impersonation of government subsidies may increase rapidly around the Lunar New Year holiday period, we hope that smartphone voice phishing detection notification services will be used to help prevent damage."

<© STARNEWS. All rights reserved. No reproduction or redistribution allowed.>

*This content was translated by AI.