As LA burns, winters get shorter and storms get stronger; the discussions around climate change are ongoing. But there\u2019s a new culprit to add to the mix: artificial intelligence. The carbon footprint caused by training AI models is much larger, and much more serious than we are hearing about. The dirty secret? They require enormous amounts of energy to train and run. There are claims that one search using Ai will use 100 times more energy than a normal browser (Sunburg, 2024). Not only is energy use a concern, but the data centres (the main powerhouses for Ai generators), are also large consumers of water. The amount of water projected to be used in 2027 is more than half of what the United Kingdom will use: \u201cthe global AI demand may be accountable for 4.2 \u2013 6.6 billion cubic meters of water withdrawal in 2027\u201d. (Islam et.al, ). Water scarcity is already a big problem, prior to the AI \u201cboom\u201d: <\/p>\n\n\n\n
\u201cApproximately 10% of the global population \u2013 around 720 million people \u2013 lived in countries with high and critical water stress levels in 2021\u201d (UN-Water, 2024). <\/p>\n\n\n\n
Not only do they require large amounts of water, but minerals and elements. Those which are often not mined sustainably (Digital Economy Report, 2024). These are used in both the data centres, as well as in the microchips which power them. Most of these elements are mined from China, and the Mountain Pass Mine in California (Rare Earth Elements, 2019). Despite having a quota system in place, China still ranks the worlds\u2019 highest production for rare earth minerals. This is concerning given their history with illegal rare earths extraction, and the current challenges with it still (Pistilli, 2025). <\/p>\n\n\n\n
While there are some statistics on this, the bigger problem is we don\u2019t know the exact numbers. There is a huge lack of transparency in how much energy usage and the environmental impacts Ai generators have. This isn\u2019t surprising as the climate crisis is hotly debated, with main perpetrators in denial of the worries. Currently, there is not reliable access to measure the greenhouse gas emitted from Ai (Dodge et al. 2022). Although concern for the environment is not new in regards to emerging technology, the mass push for Ai in many sectors from business and technology to the arts points towards these numbers becoming concerning. Some analysts in the US suggest by 2030 we will see a 20 percent jump in electricity demand, driven largely in part by A. (Gelles, 2024). <\/p>\n\n\n\n
While this is all concerning, there is hope something will be done about it. There has been legislation proposed to properly record data on both positive and negative environmental impact of Ai (Markey et Al., 2024). Hopefully in the formative years of A, proper monitoring will in fact begin, and the true magnitude of the situation can be assessed, and addressed. <\/p>\n\n\n\n
Citations<\/p>\n\n\n\n
Digital Economy Report 2024. (n.d.). https:\/\/unctad.org\/system\/files\/official-document\/der2024_en.pdf <\/p>\n\n\n\n
Dodge, J., Prewitt, T., Combes, R. T. des, Odmark, E., Schwartz, R., Strubell, E., Luccioni, A. S., Smith, N. A., DeCario, N., Buchanan, W., Jesse DodgeAllen Institute for AI, Usav. P., Taylor PrewittMicrosoft, Usav. P., Remi Tachet des CombesMicrosoft Research, C. P., Erika OdmarkMicrosoft, Usav. P., Roy SchwartzThe Hebrew University of Jerusalem, I. P., Emma StrubellCarnegie Mellon University, Usav. P., Alexandra Sasha LuccioniHugging Face, C. P., Noah A. SmithAllen Institute for AI, Usav. P., Nicole DeCarioAllen Institute for AI, Usav. P., & Will BuchananMicrosoft, Usav. P. (2022, June 20). Measuring the carbon intensity of AI in cloud instances: Proceedings of the 2022 ACM conference on fairness, accountability, and transparency<\/em>. ACM Other conferences. https:\/\/dl.acm.org\/doi\/10.1145\/3531146.3533234 <\/p>\n\n\n\n Gelles, D. (2024, July 11). A.I.\u2019s insatiable appetite for Energy<\/em>. The New York Times. https:\/\/www.nytimes.com\/2024\/07\/11\/climate\/artificial-intelligence-energy-usage.html <\/p>\n\n\n\n Markey, heinrich, eshoo, Beyer introduce legislation to investigate, measure environmental impacts of Artificial Intelligence: U.S. senator Ed Markey of Massachusetts<\/em>. Home. (2024, February 1). https:\/\/www.markey.senate.gov\/news\/press-releases\/markey-heinrich-eshoo-beyer-introduce-legislation-to-investigate-measure-environmental-impacts-of-artificial-intelligence <\/p>\n\n\n\n Pistilli, M. (2025, January 7). Top 11 countries by Rare Earth Metal production<\/em>. INN. https:\/\/investingnews.com\/daily\/resource-investing\/critical-metals-investing\/rare-earth-investing\/rare-earth-metal-production\/ <\/p>\n\n\n\n Progress on level of water stress \u2013 2024 update<\/em>. UN. (n.d.-a). https:\/\/www.unwater.org\/publications\/progress-level-water-stress-2024-update <\/p>\n\n\n\n Rare earth elements<\/em>. Semiconductor Engineering. (2019, October 26). https:\/\/semiengineering.com\/knowledge_centers\/materials\/rare-earth-elements-ree\/ <\/p>\n\n\n\n Sundberg, N. (2023, December 12). Tackling Ai\u2019s climate change problem<\/em>. MIT Sloan Management Review. https:\/\/sloanreview.mit.edu\/article\/tackling-ais-climate-change-problem\/?utm_source=twitter&utm_medium=social&utm_campaign=sm-direct <\/p>\n\n\n\n Water scarcity: UN-water<\/em>. UN. (n.d.-b). https:\/\/www.unwater.org\/water-facts\/water-scarcity <\/p>\n","protected":false},"excerpt":{"rendered":" As LA burns, winters get shorter and storms get stronger; the discussions around climate change are ongoing. But there\u2019s a new culprit to add to the mix: artificial intelligence. The carbon footprint caused by training AI models is much larger, and much more serious than we are hearing about. The dirty secret? They require enormous […]<\/p>\n","protected":false},"author":6,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"nf_dc_page":"","_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[1],"tags":[],"class_list":["post-87","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"yoast_head":"\n