Marubeni Launches AI Data Center Service Targeting 80% Power Reduction Through Spanish Tech Partnership
April 24, 2026
Japanese trading house Marubeni is set to launch a new service this month aimed at dramatically cutting the power consumption of data centers used for generative artificial intelligence, marking a significant step in the global push to make AI infrastructure more sustainable. The initiative comes as surging demand for AI computing has placed immense strain on energy grids and raised environmental concerns across the industry.
Marubeni has partnered with a Spanish startup to deploy advanced processing technology that the company says can reduce electricity usage by up to 80% compared to conventional data center operations. The service, which debuts in April 2026, targets the growing bottleneck of energy efficiency in AI workloads, where massive computational requirements have led to skyrocketing power bills and carbon footprints for hyperscale operators and enterprise users alike.
By integrating the startup's specialized hardware and software optimizations, Marubeni aims to offer clients a turnkey solution that reconfigures how AI models are trained and deployed. The technology focuses on streamlining data flow and reducing redundant computations, allowing the same AI tasks to be completed with far less electrical input. While specific financial terms of the partnership were not disclosed, the scale of the potential savings could reshape cost structures for companies building out generative AI capabilities.
The move underscores a broader industry shift toward energy-efficient AI infrastructure. Major cloud providers and data center operators have been racing to adopt liquid cooling, renewable energy sourcing, and novel chip architectures to manage power demands. Marubeni's entry into this space with a double-digit percentage efficiency gain positions the trading house as a key facilitator between technology innovators and large-scale commercial deployment in Asia and beyond.
Industry analysts note that if the 80% reduction target is validated in real-world deployments, it could accelerate the adoption of AI in regions with constrained power grids and help companies meet net-zero emissions targets. The service also highlights the growing role of trading companies like Marubeni in bridging cutting-edge European tech with Japanese and broader Asian markets.
Source: asia.nikkei
