AI Data Centers Demand More Than Copper Can Deliver
Published:Dec 27, 2025 13:00
•1 min read
•IEEE Spectrum
Analysis
This article highlights a critical bottleneck in AI data center infrastructure: the limitations of copper cables in scaling up GPU performance. As AI models grow in complexity, the need for faster and denser connections within servers becomes paramount. The article effectively explains how copper's physical constraints, particularly at high data rates, are driving the search for alternative solutions. The proposed radio-based cables offer a promising path forward, potentially addressing issues of power consumption, cable size, and reach. The focus on startups innovating in this space suggests a dynamic and rapidly evolving landscape. The article's inclusion in a "Top Tech 2026" report underscores the significance of this challenge and the potential impact of new technologies on the future of AI infrastructure.
Key Takeaways
- •Copper cables are becoming a bottleneck for scaling AI data centers.
- •Radio-based cables offer a potential solution to overcome copper's limitations.
- •Startups are actively developing and integrating radio cables with GPUs.
Reference
“How fast you can train gigantic new AI models boils down to two words: up and out.”