AT&T and Ericsson demonstrate AI-native Cloud RAN

AT&T and Ericsson completed a Cloud Radio Access Network (RAN) test that demonstrated Ericsson’s AI-native Link Adaptation on a Cloud RAN stack powered by Intel Xeon 6 SoC, and the trial showed improvements in spectral efficiency and network responsiveness.

The test RAN on AT&T’s licensed frequency bands and the companies described that it provided a technical template for scaling AI-native RAN functions into AT&T’s cloud infrastructure and for operating Machine Learning (ML) natively within radio environments. Dell’Oro Group estimated Cloud RAN accounted for approximately 5% to 10% of total global RAN market revenues in 2025.

AI-native Link Adaptation used a learning algorithm that continuously assessed channel state and interference to determine the modulation and coding scheme for each transmission interval. The model generated real-time predictions of link quality and adjusted data rates to increase throughput and spectral efficiency compared with conventional rule-based link adaptation.

The proof-of-concept RAN Ericsson’s disaggregated, containerized RAN software within AT&T’s target Cloud RAN configuration on commercial off-the-shelf hardware and Intel Xeon 6 SoC. The trial progressed from basic call functionality to validation of feature-rich network behavior; early results recorded throughput gains up to 20% and measurable improvements in spectral efficiency. Ericsson and Intel benchmarked Artificial Intelligence (AI) inference models and demonstrated performance scalability and energy efficiency on general-purpose compute nodes.

Rob Soni, Vice President, RAN Technology at AT&T, said: “AT&T is leading the charge toward an open, intelligent, and scalable network future by advancing Open RAN and Cloud RAN with AI-native capabilities at their core. This demo highlights how AI capabilities, powered by our next-generation Cloud RAN platform, can be deployed seamlessly to drive innovation and deliver superior customer experiences.”

The companies described plans to scale AI-native RAN functions into AT&T’s cloud infrastructure and to explore ML operating natively within radio environments for future 5G and 6G network evolution.