How SK Telecom uses AI

SK Telecom says it blocked about 1.1 billion attempted telecom fraud cases in 2025, 35 percent more than a year earlier. It introduced an AI model based on call pattern analysis that detects suspected voice phishing numbers not yet reported to authorities, and applied its in house ScamVanguard AI fraud detection to spam filtering in its PASS app and the A.Dot Phone. The announcement states 850 million text messages blocked.

SK Telecom built an AI Customer Service Support System on its own telecommunications specialized large language model (Telco LLM) and large multimodal model (LMM), trained with input from dozens of customer service experts over about a year. The system gives agents an AI Knowledge Search Assistant that answers natural language questions instead of manual lookup, an LMM powered Intelligent Document Processing feature that classifies customer submitted documents, and an Automated Post-Processing feature (planned at the time of the announcement) to summarize consultation outcomes. It completed a month long beta test starting in mid October before this announcement.

SK Telecom built an image analysis model in house that reads drone photos of cell towers, which reach up to 75 meters, and judges the condition of bolts and nuts. Before, inspectors had to climb the towers or study about 100 images per tower by eye. SKT says the model cut the time for that review by 95% and made the analysis more consistent, and plans to add wind pressure and inclination checks and a link to its safety management system.