LGU+ Applies Ixi Spam Filter System
Filtering Accuracy at 95% Since Initial Deployment
Management Policies for Similar Investment, Gambling Types Also Implemented
LG Uplus announced on the 29th that it has applied the 'ixi spam filter' system, which uses artificial intelligence (AI) to filter spam messages on its mobile communication network, to prevent fraud damage caused by illegal spam messages.
Illegal spam messages refer to text messages sent in bulk without the consent of communication service subscribers for commercial or malicious purposes. To prevent fraud damage caused by illegal spam messages, it is necessary to detect and block specific keywords or patterns in the text messages that are difficult for humans to identify.
LG Uplus developed the AI-based ixi spam filter to reduce the sending of illegal spam messages. This system learns from spam report data received from the Korea Internet & Security Agency (KISA) through an AI model and blocks spam messages before users receive them. From the initial implementation, it has raised the filtering accuracy of illegal spam messages to over 95%, and it detects new types of spam messages through real-time monitoring. The plan is to continuously enhance its response capabilities by learning the latest illegal spam messages daily.
LG Uplus expanded the types of mass-sent malicious messages under management from the existing categories of ▲illegal loans ▲smishing ▲gambling ▲adult content ▲illegal pharmaceuticals to include 'similar investment' such as stock leading chat rooms. Since last month, the company has been operating a sanction policy that blocks the sender's number and requests an explanation from the message sending operator if even one spam message of these types is sent.
Additionally, LG Uplus has strengthened supervision of resellers who send large volumes of text messages using its own network. For resellers with no usage in the past six months, a policy limiting the number of messages they can send to once per month is being enforced, and a blacklist of illegal spam sending history is shared with related organizations.
As a result of these policies, the number of spam message reports last month decreased by 10% compared to the monthly average in the first half of the year.
Meanwhile, LG Uplus plans to equip the upcoming call agent 'ixi-O', scheduled for release next month, with an on-device AI-based voice phishing detection solution. Through the ixi spam filter and the voice phishing detection solution, users will be protected on both text messages and calls.
Hong Gwanhee, Head of the LG Uplus Cybersecurity Center (CISO, Executive Director), said, "We will safely protect users from illegal spam messages through AI technology and establish a rapid and accurate response system," adding, "We will continue to increase the utilization of AI and security technologies to further enhance service reliability."
© The Asia Business Daily(www.asiae.co.kr). All rights reserved.
