Topic · Concept

AI datacenters

AI datacenters are the facilities, chips and power infrastructure used to train and run AI models. Their scale is now measured in gigawatts of electricity rather than server counts, which makes energy supply, chip availability and capital spending central to the AI story.

Follow via RSS Get AI datacenters alerts4 stories · updated Oct 9

Fact file

Data centre electricity use, 2024~415 TWh, ~1.5% of global electricity (IEA estimate)

Projection, 2030~945 TWh (IEA base case)

IEA report publishedApril 10, 2025

IEA update, 2025 use~485 TWh in 2025 (+17%); ~950 TWh by 2030 (IEA "Key Questions on Energy and AI", Apr 16, 2026)

Stargate announcementJan 21, 2025; $500B planned over four years (company plan); equity funders SoftBank, OpenAI, Oracle, MGX

Facts verified Oct 9

Latest on AI datacenters

Oct 9Business · News4 minOpenAI reportedly seeks $30 billion at a $1.4 trillion valuation as it deepens Atlassian deal Oct 9Hardware · News4 minNvidia's Vera Rubin reaches customer production as AMD buys an AI lab Oct 9Hardware · Analysis8 minThe gigawatt datacenter race: what's announced, what's running, and what's holding it back Oct 9Business · Explainer8 minAnthropic explained: history, products and how it makes money

Nothing in this format on this page — try another filter or the next page.

Timeline

APR 2025IEA report published