NVIDIA Partners with Global Operators to Define AI-Native 6G, with Storage as a Key Foundation

At Mobile World Congress 2026, NVIDIA announced a joint commitment with leading global telecommunications companies—including Booz Allen, BT Group, Cisco, Deutsche Telekom, Ericsson, Nokia, SK Telecom, SoftBank, and T-Mobile—to build next-generation 6G wireless networks on an open, trusted, AI-native software-defined platform. "Artificial intelligence is redefining computing, driving the largest-scale infrastructure build in human history—and telecommunications is the next frontier," said Jensen Huang, founder and CEO of NVIDIA. "NVIDIA is collaborating with an alliance of global industry leaders to create AI-RAN, transforming global telecom networks into global AI infrastructure."

Collaboration Details: AI-RAN Technology Roadmap and Open Ecosystem

The core of this collaboration is AI-RAN technology. By embedding AI into the radio access network, edge, and core, 6G networks will support integrated sensing and communications while possessing intelligent decision-making capabilities. NVIDIA is not only advancing technology standardization but has also reached new AI-RAN commercialization partnerships with T-Mobile and SoftBank, moving the technology from the lab toward deployment. On the software front, NVIDIA has released a large language model for telecommunications based on the Nemotron framework, along with agent-building guidelines for network operations center workflows, covering scenarios such as energy efficiency and multi-agent orchestration.

Development Direction: Physical AI Foundation and Autonomous Network Vision

6G will serve as the foundation for physical AI, enabling billions of autonomous machines, vehicles, and robots—imposing higher demands on network security and trust. NVIDIA has proposed the long-term goal of "autonomous networks"—where telecom networks can self-manage like intelligent agents, requiring large models and inference systems specifically designed for telecommunications. T-Mobile's CEO stated that 6G will become the nervous system of the digital economy, driving automation and the development of intelligent industries. According to plans, 6G is expected to become commercially available around 2030, with trials starting as early as 2028. NVIDIA is attracting diverse ecosystem participants—from operators to startups—through its open, programmable platform.

The Critical Supporting Role of Storage

Within the AI-native 6G blueprint, storage—as the core carrier of data and models—will provide critical support for the network's continuous learning and intelligent evolution. 6G needs to process massive amounts of sensing data, run complex AI models, and support continuous iteration. This requires not only high-performance storage to host petabyte-scale data lakes but also ultra-low latency data access capabilities for AI training and inference. Storage infrastructure will become the "data heart" of this transformation, ensuring that network agents can call upon historical data in real-time, optimize configurations, and ultimately support the vision of "autonomous networks."