The useful question is what changes for users, developers or buyers, and whether the announcement stays industry context or becomes something people can actually use.

Samsung Electronics and KDDI concluded a trial of the AI-powered RAN Speed Optimizer (RSO) on KDDI’s commercial 5G Standalone network in Japan. The trial, conducted from late 2025 across Tokyo and surrounding regions, tested the solution in urban, suburban, and rural environments using 100 MHz of 3.7 GHz TDD spectrum. The test aimed to evaluate AI’s ability to enhance network performance in real-world conditions, with results showing measurable improvements during peak usage periods.

During the trial, Samsung’s RSO delivered an average 31% increase in 5G downlink throughput across the test area, with peak gains of 52% in dense urban zones. The solution uses AI to optimize parameters for each cell individually, moving beyond traditional cluster-level tuning that applies uniform settings. This approach allows the network to adapt dynamically to varying traffic and environmental conditions, reducing the need for manual adjustments.

The RSO is part of Samsung’s CognitiV Network Operations Suite (NOS), a collection of AI-driven automation tools designed to improve network efficiency and reduce operational costs. By automating parameter adjustments, the system enables operators to respond more quickly to changing network demands, improving reliability and performance for end users. The trial demonstrated the potential for AI to streamline network management in commercial environments.

KDDI and Samsung plan to expand their collaboration on AI-based optimization technologies following the trial’s success. The companies highlighted the role of their long-standing partnership in virtualized network deployments as a foundation for advancing toward AI-native network operations. The trial underscored the practical benefits of AI-driven network optimization for both operators and subscribers.