Ericsson has introduced a new software subscription designed to bring telco-grade artificial intelligence models directly into radio access networks. The new offering allows mobile operators to improve 5G network performance, automation, and energy efficiency without deploying additional hardware. The initiative marks another step in Ericsson’s broader vision for AI-native telecommunications networks.
The subscription, called AI in RAN, embeds AI models into baseband and radio equipment. As a result, operators can leverage real-time AI capabilities within existing network infrastructure while preparing for future AI-native network architectures.
AI Enhances 5G Performance and Efficiency
According to Ericsson, the software uses telco-grade AI models optimized for real-time operation within radio access networks. Moreover, the system supports continuous learning and agentic AI capabilities, enabling more advanced network management and automation.
The solution works across Ericsson 5G Advanced deployments, including purpose-built and Cloud RAN environments. Furthermore, Ericsson leverages its proprietary radio silicon and latest RAN computing technologies to deploy AI models where they can deliver the greatest impact.
Initial features include AI-powered scheduling for link adaptation, beamforming optimization, and multi-layer coordination. Consequently, operators can improve network resource allocation and overall user experience.
Ericsson stated that trials conducted across more than 15 deployments worldwide delivered measurable gains. The company reported up to 20% higher downlink throughput and up to 10% better spectral efficiency. In addition, the technology supported twice as many high-traffic users while achieving up to 95% coverage prediction accuracy.
Industry Moves Toward AI-Native Networks
Mårten Lerner, Head of Networks Strategy and Product Management at Ericsson, said the company is “redefining what’s possible in mobile networks by bringing powerful AI capabilities to service providers,” adding that “we are taking a major step toward AI-native networks.”
Several telecom operators, including SoftBank, Bell, SK Telecom, and Rogers, highlighted the potential benefits of AI-powered network optimization, energy savings, and automation. Therefore, industry support suggests growing confidence in AI-driven network operations.
Joe Madden, Principal Analyst at Mobile Experts, added: “With a software upgrade, operators can squeeze more capacity, better observability, and more accurate location-based services out of the 5G network they bought years ago”.
As telecom providers face rising traffic demands and increasingly complex network environments, AI is becoming a critical component of future infrastructure strategies. Consequently, Ericsson’s new subscription model offers operators a practical path toward AI-enabled networks without requiring immediate hardware upgrades.








