SKT: "AI Detects Voice Phishing Even During Calls"

[Financial News] SK Telecom announced on the 1st that it has newly added the call security feature 'AI Voice Phishing Detection' to its AI phone service, Adot Phone.
AI Voice Phishing Detection analyzes conversations in real time during calls and immediately notifies users if a call is suspected to be voice phishing. SK Telecom expects this feature to help protect users from increasingly sophisticated voice phishing crimes and provide a safer calling environment.
The AI model determines suspicion by comprehensively analyzing various factors such as the presence of suspicious keywords and conversation patterns in real time during calls. However, calls from numbers saved in the user's contacts or official business numbers registered in Adot Phone's Business Contacts are excluded from analysis.
Analysis results are categorized into two levels—'Suspicious' and 'Danger'—depending on severity. Users are immediately alerted during calls through warning pop-ups, notification sounds, or vibrations.
If the user does not select 'Not Voice Phishing' on the warning pop-up, the number is labeled as 'Phishing Detected' after the call ends. This warning label can be checked in screens such as 'Recent Records' and 'Search' within the Adot Phone app.
When a user attempts to call a number marked with a warning label, Adot Phone displays a warning pop-up to confirm whether the call should proceed. If a call is received from a suspicious number, Adot Phone displays warning messages such as 'Detected as Voice Phishing' on the incoming call screen to alert the user.
This feature is based on On-Device AI technology, processing the entire sequence from call analysis to warning notifications directly on the device. The AI model analyzes call content in real time on the device, continuously detecting signs of voice phishing.
SK Telecom explained that, because this approach does not route call data through servers or involve separate data storage or deletion processes, it enables secure detection without concerns about information leakage.
AI Voice Phishing Detection is available through the Adot Phone app on the Android operating system (AOS) and via the 'Phone' menu in the Adot app on iOS. To use the feature, users must activate AI Voice Phishing Detection in the 'AI Security' menu under settings.
The 'AI Security' menu consolidates key security features of Adot Phone, such as 'AI Voice Phishing Detection' and 'AI Safe Call Block,' allowing users to easily check and configure them. The menu also includes the newly added 'Call Interception Detection' feature, which warns users if a malicious app attempts to redirect outgoing calls to another number.
Additionally, SK Telecom has implemented various measures to ensure safe calling experiences for customers, such as enabling users to instantly access contact information for relevant reporting agencies by entering keywords like 'voice phishing' or 'smishing' in the 'Search' screen.
Meanwhile, SKT is working to strengthen its security posture from prevention to recurrence prevention through robust customer protection measures, the launch of security solutions, and public-private partnerships.
Since last November, SK Telecom has been offering the global mobile security solution 'Zimperium' free of charge to customers for one year. In addition, the Korean National Police Agency (KNPA) and the three major mobile carriers in South Korea (SK Telecom, KT, and LG Uplus) have collaborated to implement the 'Emergency Blocking System,' which blocks phone numbers used in crimes such as voice phishing within 10 minutes. This system has been in effect since the 24th.
mkchang@fnnews.com Jang Min-kwon Reporter