Samsung, KT Successfully Validate AI-Based RAN Optimization on Commercial Network

Wedoany.com Report-Dec.12, Samsung Electronics and KT have successfully validated AI-based radio access network (AI-RAN) optimization technology on KT's live commercial 5G network, achieving the world's first real-network demonstration of user-level AI-driven connectivity improvements.

Following simulation verification completed in June 2025, the companies conducted extensive field testing in Seongnam, Gyeonggi Province, involving approximately 18,000 real subscribers across diverse urban environments. The trial focused on users who had repeatedly experienced connection drops due to movement through weak-coverage zones or high-speed travel.

Traditional networks apply uniform radio parameters to all devices connected to a base station. The validated AI-RAN solution instead uses artificial intelligence to analyse each user's real-time radio conditions, movement patterns and historical disruption data. It then automatically applies individualized optimal configurations, proactively preventing disconnections before they occur.

Results showed a substantial reduction in connection failures for previously affected users and an overall improvement in service stability across the test area.

JinGuk Jeong, Executive Vice President and Head of the Advanced Communications Research Center at Samsung Research, said: "This is a major milestone that shows how AI can improve user experience on real-world commercial networks."

Jong-Sik Lee, Executive Vice President and Head of Future Network Laboratory at KT, added: "This achievement demonstrates that AI can transform network operations around users."

The technology is regarded as a foundational element for 6G networks, where explosive data growth and diverse communication scenarios will demand highly adaptive, user-centric performance. Samsung and KT plan to broaden validation to additional commercial regions and continue joint development of next-generation AI-native network capabilities.

The successful live-network trial confirms that AI-RAN can deliver measurable quality-of-experience gains without requiring new device hardware, paving the way for wider adoption by operators seeking to differentiate service quality in competitive markets.