LG Uplus<\/a> built a voice phishing prevention system that covers the full customer journey. The approach spans three sequential stages: monitoring, crime response, and emergency intervention. It also combines AI detection with real-time network monitoring and direct cooperation with law enforcement.<\/p>\n\n\n\n
\n\n\n\nSolution<\/h2>\n\n\n\n
LG Uplus\u2019s voice phishing prevention approach covers three sequential stages, each targeting a different point in the attack chain.<\/p>\n\n\n\n
Three-stage voice phishing prevention framework<\/h4>\n\n\n\n
The monitoring stage runs 24\/7 through the Customer Damage Prevention Analysis System. This blocks spam messages and prevents subscriber access to malicious URLs in real time. LG Uplus is also the only mobile operator in South Korea that tracks criminal command-and-control servers. Criminal organisations use these servers to distribute malicious apps to potential victims. Once installed, these apps intercept calls and redirect 112 emergency calls to criminals. They also activate the victim\u2019s camera for surveillance. LG Uplus identifies subscribers who have accessed these servers, blocks the connection, and shares the data with police.<\/p>\n\n\n\n
The crime response stage centres on ixi-O, LG Uplus\u2019s AI call agent. ixi-O analyses live call audio and warns the subscriber with a voice and visual alert. It also uses Anti Deepvoice to identify voices that criminals synthesise to impersonate family members or officials. Since its launch in November 2024, ixi-O has flagged an average of 2,000 suspicious calls per month. In February 2025, LG Uplus also upgraded its AI-based spam blocking system. Consequently, the volume of blocked spam messages containing malicious URLs rose by approximately 1.4 times within five months.<\/p>\n\n\n\n
Emergency response and cross-sector partnerships<\/h4>\n\n\n\n
When LG Uplus confirms that a subscriber has installed a malicious app, it sends an Alert Talk via KakaoTalk. KakaoTalk avoids the SMS channel, which criminals on infected devices can intercept and delete. Subscribers who receive an alert can seek immediate help at more than 1,800 LG Uplus stores. They can also contact nearby police officers. Since launch in June 2025, the service alerted approximately 3,000 subscribers within its first four weeks.<\/p>\n\n\n\n
LG Uplus also established a formal on-site cooperation arrangement with the Seoul Metropolitan Police Agency. This is the first arrangement of its kind in the South Korean telco industry. Company staff accompany police during visits to potential victims and carry out on-site malicious app detection. LG Uplus also signed a Telecom-Finance Fraud Prevention Agreement with Woori Bank. This created a shared intelligence channel between the telecoms and financial sectors.<\/p>\n\n\n\n
\n\n\n\nImpact<\/strong><\/h2>\n\n\n\nVoice phishing prevention outcomes and scale<\/h4>\n\n\n\n
LG Uplus\u2019s spam blocking service has blocked 250 million bait-type spam messages since launch. It has also detected 880,000 smishing attempts over the same period. These efforts prevented an estimated KRW200 billion in voice phishing damages over six months. Within one month of launching Anti Deepvoice, LG Uplus detected over 5,500 manipulated voice calls. Prevented damages from that technology alone reached an estimated KRW290 billion.<\/p>\n\n\n\n
After the February 2025 upgrade, blocked spam messages rose by approximately 1.4 times within five months. Customer reports of received spam also fell by about 65% in the same period. The AI automatic blocking rate increased eightfold as a result.<\/p>\n\n\n\n
In Q2 2025, LG Uplus\u2019s app tracking contributed to roughly 23% of all NPA voice phishing cases. The Ministry of Science and ICT also recognised the company\u2019s security capabilities. LG Uplus received a Minister\u2019s Award at South Korea\u2019s Information Security Awards.<\/p>\n\n\n\n
LG Uplus invested KRW82.8 billion in information security in 2024. The company also plans to invest KRW700 billion in the field over the next five years. Overall, the LG Uplus model shows that voice phishing prevention can scale through AI, network monitoring, and sector partnerships.<\/p>\n\n\n\n
\n\n\n\nCost:<\/strong> Low<\/p>\n\n\n\nParticipating organisations:<\/strong><\/strong> LG Uplus<\/p>\n\n\n\n <\/div>\n \n