LG Uplus Enhances AI-Based Spam Blocking... Number of Blocked Messages Increases by 1.4 Times

Customized AI Filtering... Rapid Detection
Analyzing and Proactively Blocking Bait Messages
Spam Message Reports Decreased by 65%

On July 16, LG Uplus announced that, as a result of enhancing its artificial intelligence (AI)-based spam blocking system, the number of blocked spam messages increased by approximately 1.4 times in just five months. As 'bait messages' sent by voice phishing and smishing criminals to deceive victims are becoming increasingly sophisticated, it is analyzed that customized AI filtering has proven effective.

As a result of LG Uplus enhancing its artificial intelligence (AI)-based spam blocking system, the number of blocked spam messages increased by approximately 1.4 times in six months. Photo shows LG Uplus employees operating the spam blocking system. Photo by LG Uplus

Bait messages are text messages sent by voice phishing and smishing criminals who impersonate public institutions, financial companies, or delivery services to lure recipients into clicking malicious internet URLs, or who disguise themselves as family members to induce remittance. As the criminals' methods become more sophisticated, it is becoming increasingly difficult to distinguish between legitimate messages and bait messages based solely on their content.

In response, LG Uplus has introduced a solution to its customer damage prevention analysis system, an AI-based internal and external data integration analysis and response framework, that can identify and proactively block bait messages.

LG Uplus continuously updates the phrases that need to be blocked by training the AI with accumulated spam message data and the latest trends. At the same time, it analyzes whether the sender's number has been manipulated or if the message was sent from overseas to determine whether to block it. For example, if a message containing the keyword 'card delivery' is analyzed and it is found that the sender's number is manipulated and the message was sent from overseas, it is blocked as spam.

This solution was first implemented in February. Comparing June to January, before the solution was introduced, the number of blocked spam messages increased by approximately 1.4 times, and the number of actual customer reports of received spam messages decreased by about 65% during the same period.

As time passes, the performance of the AI continues to improve, resulting in an eightfold increase in the AI's real-time automatic blocking rate. With more accurate and faster detection, customer protection has been strengthened, and the inconvenience caused by receiving unnecessary spam messages has been reduced.

LG Uplus plans to continue strengthening its spam blocking solution to prevent customer damage. The company will add new features to detect spam message patterns more quickly and will further enhance its AI to increase the probability of spam prediction.