AI server consumes raw materials
From chip manufacturing and the expansion and operation of data centers to the training and everyday use of AI models, these systems increasingly demand vast and growing quantities of electricity, water and raw materials. Older "brownfield" data centers were designed for server racks consuming between 5 and 15 kilowatts (kW) of power. Founded at the Massachusetts Institute of Technology in 1899, MIT Technology Review is a world-renowned, independent media company whose insight, analysis, reviews, interviews and live events explain the newest technologies and their commercial, social and political impact. AI's rapid expansion also drives higher water usage, emissions, and e-waste, raising urgent sustainability concerns, according to Mahmut Kandemir, a distinguished professor in the Department of Computer. AI environment statistics for 2026 paint a stark picture: artificial intelligence could soon consume nearly half of all global data center electricity, overtaking even Bitcoin mining in energy use. The hidden cost behind every ChatGPT prompt, AI search, or image generation is no longer abstract;. A Google Tensor Processing Unit, an example of application-specific integrated circuits used for AI.
Read More