Hyperscale data centers are now powering AI models with a revolutionary architecture—at a staggering energy cost.

Technology Review, published by MIT, is a leading source of news, analysis, and commentary on emerging technologies, scientific advancements, and their societal impacts. Covering topics such as artificial intelligence, biotechnology, renewable energy, and more, Technology Review provides  reporting and  articles on the intersection of technology and society. Readers can stay informed about the latest breakthroughs and trends shaping the future of technology.

MIT Technology Review

Hyperscale AI data centers are emerging as massive facilities housing hundreds of thousands of GPUs (like Nvidia's H100s) connected by fiber-optic cables to train and run large language models. Tech giants like OpenAI, Google, Microsoft, and Meta are investing hundreds of billions in this infrastructure. These facilities consume over a gigawatt of electricity—enough to power entire cities—with over half coming from fossil fuels. The extreme heat from densely packed chips requires liquid cooling systems, including cold water plates or cooling fluid baths. Communities hosting these facilities face rising energy costs, water shortages, noise pollution, and air quality issues as the AI industry scales rapidly.

Hyperscale AI data centers: 10 Breakthrough Technologies 2026