GoodVision AI has signed a strategic partnership with AI Storm to build and scale AI infrastructure in Japan, starting with a flagship AI Factory in Fukushima. The project marks GoodVision’s first deployment in the country and positions Japan as a key market in its broader global expansion strategy.
The first phase involves a 2 MW liquid-cooled facility equipped with 72 NVIDIA B300 servers and more than 500 GPUs dedicated to AI inference workloads. The site is expected to be operational within three months and will serve as the blueprint for future deployments across the country.
The companies are targeting 20 MW of installed AI capacity within the next year and plan to reach 100 MW within three years through a network of distributed AI Factory sites. More than 50 potential locations have already been secured across Tokyo and nearby regions to support the rollout.
Also Read: Data Centers and Batteries Are Becoming the Same Story
The strategy shows a growing change in how people plan AI infrastructure. Rather than leaning only on a few giant centralized data centers, many providers are starting to look into regional setups and edge deployments. These options put the compute capacity nearer to businesses, research institutions, and places with high demand, which kind of makes sense, because the distance matters.
GoodVision’s platform combines GPU integration, power management, liquid cooling, and operational support into a single infrastructure stack. The company also plans to integrate its Smart Routing Engine technology to direct AI workloads across available models and computing resources based on performance, cost, latency, and data requirements.


