Pranav Hotkar
Tokyo, Japan - March 2, 2026 - SoftBank Corp. and Mitsubishi Heavy Industries have launched a strategic collaboration to advance edge computing solutions that integrate artificial-intelligence processing with network infrastructure, marking a step forward in edge data center development and AI-native connectivity.
The partnership centers on field trials using SoftBank’s AITRAS AI-RAN product, a radio access network technology enhanced with AI, within an edge data center environment established by MHI at its Yokohama Hardtech Hub. Both companies said the initiative aims to validate high-performance, secure edge AI processing in coordination with low-latency network capabilities.
Under the agreement, SoftBank will deploy AITRAS at MHI’s demonstration site to test how AI-enabled RAN functions can be combined with localized compute to handle tasks such as real-time inference and secure data processing, separated from public networks. MHI’s edge infrastructure will host the trial, leveraging its modular edge computing and power systems designed for industrial workloads.
“This collaboration enables us to explore new models for delivering AI-driven services at the network edge,” SoftBank said in its press release, adding that the field trials will focus on scenarios where rapid, secure, and stable communication with local compute is critical.
MHI said the initiative builds on its broader efforts to expand industrial digital infrastructure, including its recently unveiled DIAVAULT edge data center platform, which integrates computing, energy and cooling systems for secure, high-performance edge processing. MHI described the partnership as an opportunity to combine advanced network technologies with edge compute capabilities that can support applications in manufacturing, logistics, telecommunications, and other industries requiring real-time analytics and AI inference.
Edge computing is increasingly seen as essential as enterprises and network operators look to reduce latency and increase reliability for AI workloads that cannot be efficiently processed in centralized cloud facilities. By situating compute resources closer to data sources and integrating them with intelligent networking, the SoftBank-MHI collaboration is designed to meet demand for use cases such as autonomous systems, industrial automation, and localized analytics.
The field trial, located at MHI’s Yokohama Hardtech Hub, will test secure, high-speed AI processing using edge data center infrastructure alongside AI-RAN network functions. Both companies said insights from the tests will inform future deployments and product development strategies as they work to scale AI-ready edge computing solutions.
The initiative highlights a broader trend in which telecommunications carriers and industrial infrastructure providers increasingly collaborate to integrate network intelligence with edge compute, reflecting the growing importance of distributed AI and real-time data processing at the network edge.
