Subscribe to Our Newsletter

Success! Now Check Your Email

To complete Subscribe, click the confirmation link in your inbox. If it doesn’t arrive within 3 minutes, check your spam folder.

Ok, Thanks
KT Develops AI Technology to Combat Voice Phishing
Source: KT

KT Develops AI Technology to Combat Voice Phishing

Philip Lee profile image
by Philip Lee

Seoul, South Korea—KT Corp (030200. KS) announced on Tuesday the development of two artificial intelligence (AI) technologies to protect customers from voice phishing scams.

The telecom giant has created an 'AI Voice Phishing Detection Agent' and an 'AI Suspicious Number Alert Service' to enhance real-time detection of fraudulent calls.

The AI Voice Phishing Detection Agent utilizes real-time speech-to-text conversion and a small language model (sLM) to identify context related to financial fraud during phone conversations swiftly. 

It alerts customers immediately if suspicious circumstances are detected.

KT's new technology improves upon existing keyword-based detection methods by comprehensively analyzing various risk situations, such as requests for identification, passwords, or money. 

The system operates on-device, preventing external data transmission and enhancing privacy protection.

The National Forensic Service provided anonymized voice phishing scripts to aid development.

 KT plans to continue collaboration with the agency to improve its AI detection capabilities further.

The AI Voice Phishing Detection Agent is beta testing with KT's internal experts. 

It is scheduled for official release in October through the spam-blocking app "Who Who," which has over 6 million users.

The AI Suspicious Number Alert Service uses an AI classification model to pre-emptively detect and warn customers about potentially fraudulent phone numbers. 

This service is undergoing employee testing and is being prepared for broader release as an add-on service for KT customers.

KT has also implemented an "Emergency Network Blocking Service" since July, working with law enforcement to block numbers used in voice phishing crimes in real time.

Philip Lee profile image
by Philip Lee

Subscribe to The Pickool

Success! Now Check Your Email

To complete Subscribe, click the confirmation link in your inbox. If it doesn’t arrive within 3 minutes, check your spam folder.

Ok, Thanks

Read More