How much energy does AI actually consume, and how does it compare to other major industries?
TechnologyAI Energy
The sources do not provide specific quantitative data on AI's total energy consumption, such as gigawatt-hours or percentages of global energy use. Instead, they highlight concerns about a significant surge in energy demand driven by AI infrastructure expansion, including data centers, which is threatening higher electricity costs for residential users and straining power grids [3][5][12]. OpenAI CEO Sam Altman defends AI's energy use by comparing it to the substantial energy required to raise or train humans, arguing that AI is more efficient in the long term and could even enhance overall energy efficiency through innovations [1][2][6][7][8]. No direct comparisons to other major industries like manufacturing or aviation are available in the sources.
Sources
- OpenAI CEO Defends AI Energy Efficiency — Top Daily Headlines
- Humans use a lot of energy, too — The Deep View
- AI Boom Sparks Surge in Energy Demand, Threatening Higher Electricity Costs — Daily Brew
- AI Infrastructure Costs Skyrocket — The Register
- From AI Winters to Energy Chokepoints — Daily Brew
- Sam Altman gets defensive about AI’s massive electricity usage: ‘It also takes a lot of energy to train a human’ — fortune
- Assessing the global impact of artificial intelligence on energy — RePEc
- OpenAI CEO defends AI energy usage by comparing it to human training — Axios AI+
- Compute Costs Dominate AI Company Expenses — Exponential View
- AI Investment Gains May Be Energy's Loss — Axios AI+
- Anthropic Covers Data Center Energy Costs — The Rundown AI
- Advanced economies face AI-Energy trade-off — Devdiscourse
- Energy demand from AI – Energy and AI – Analysis - IEA — IEA
- As Use of A.I. Soars, So Does the Energy and Water It Requires - Yale E360 — Yale e360
- AI and energy: Will AI reduce emissions or increase power demand? | World Economic Forum — World Economic Forum
Related questions
- →How should organisations factor AI energy consumption into their sustainability and ESG reporting?
- →What innovations in energy efficiency are most likely to reduce AI's environmental impact at scale?
- →How are hyperscalers sourcing energy for AI data centres, and what pressure does this place on grids and energy markets?
- →What is the carbon footprint of training a large frontier model — and of running it at scale?