Salta al menu principale di navigazione Salta al contenuto principale Salta al piè di pagina del sito

Geografia e intelligenza artificiale: spazialità, reti, narrazioni nella società algoritmica

N. 1 (2026)

Intelligenza artificiale e domanda di energia: una prospettiva geografica

  • Domenico de Vincenzo
DOI
https://doi.org/10.3280/rgioa1-2026oa22422
Inviata
8 aprile 2026
Pubblicato
11-05-2026

Abstract

L’intelligenza artificiale è strettamente legata alla questione energetica. Le sue infrastrutture fondamentali, i data center, richiedono grandi quantità di energia. Questa forte dipendenza da fonti energetiche concentrate nello spazio rende l’IA una tecnologia territorialmente selettiva e, al tempo stesso, ambivalente: può contribuire all’efficienza energetica e alla transizione ecologica, ma può anche generare nuovi squilibri socio-ambientali. La crescente domanda globale di energia da parte dei data center, che secondo l’International Energy Agency potrebbe più che raddoppiare entro il 2030, incide in modo significativo sull’organizzazione spaziale delle infrastrutture digitali. I criteri localizzativi dei data center variano in base alla funzione svolta nell’ambito dell’intelligenza artificiale, distinguendo due fasi: training e inferenza. La fase di training richiede grandi quantità di energia ma non impone vincoli stringenti sulla latenza della rete. Per questo motivo, i data center dedicati a questa fase tendono a essere collocati in aree periferiche. Nella fase di inferenza, che comporta l’interazione in tempo reale tra l’intelligenza artificiale e gli utenti, la priorità diventa la velocità e l’affidabilità della rete, fattori che spingono la localizzazione dei data center in prossimità dei grandi centri urbani. Queste scelte generano ingiustizie ambientali e energetiche, in cui alcuni territori si specializzano nell’attrazione di grandi infrastrutture digitali (spesso ad alto impatto ambientale), subendone gli effetti negativi. La presenza di data center hyperscale può infatti innescare conflitti legati all’uso delle risorse energetiche e ambientali, mettendo in luce il ruolo sempre più centrale del territorio nelle dinamiche dell’economia digitale.

Riferimenti bibliografici

  1. Andrae A.S.G., Edler T. (2015). On global electricity usage of communication technology: Trends to 2030. Challenges, 6(1): 117-157. DOI: 10.3390/challe6010117.
  2. Bolón-Canedo V., Morán-Fernández L., Cancela B., Alonso-Betanzos A. (12024). A review of green artificial intelligence: Towards a more sustainable future. Neurocomputing, 599. DOI: 10.1016/j.neucom.2024.128096.
  3. Bourzac K. (2024). Fixing AI’s energy crisis. Nature.com, 17 ottobre 2025. Testo disponibile al sito: www.nature.com/articles/d41586-024-03408-z (consultato il 12 dicembre 2025).
  4. Bridge G., Gailing L. (2020). New energy spaces: Towards a geographical political economy of energy transition. Environment and Planning A: Economy and Space, 52(6): 1037-1050. DOI: 10.1177/0308518X2093957
  5. Central Statistics Office (2024). Data Centres Metered Electricity Consumption 2023. Ireland Central Statistics Office. Testo disponibile al sito: cso.ie/en/releasesandpublications/ep/p-dcmec/datacentresmeteredelectricityconsumption2023/keyfindings/ (consultato il 23 luglio 2025).
  6. Cleveland-Peck P. (2025). Microsoft Feels the Heat as ‘Carbon Negative’ Goal Looms Nearer. The Wall Street Journal, wsj.com, 29 maggio.
  7. Commissione Europea (2023). Regolamento delegato della Commissione n. 1087 del 10/2/2023.
  8. Cowls J., Tsamados A., Taddeo M., Floridi L. (2023). The AI gambit: Leveraging artificial intelligence to combat climate change - Opportunities, challenges, and recommendations. AI & Society, 38: 283-307. DOI: 10.1007/s00146-021-01294-x
  9. Davis M. (2024). Impact of Pollution on Marginalized Communities: Case Study of Omaha, Nebraska. Lincoln: University of Nebraska.
  10. De Falco S. (2024). Urban dimension of crypto mines dynamics: A spatial approach. USA focus. Archivio di Studi Urbani e Regionali, 139: 115-136. DOI: 10.3280/ASUR2024-139006.
  11. de Roucy-Rochegonde L., Buffard A. (2025). AI, Data Centers and Energy Demand: Reassessing and Exploring the Trends. Parigi: Ifri.
  12. de Vries A. (2023). The growing energy footprint of artificial intelligence. Joule, 7(10): 2191-2194.
  13. Engel A., Posner D., Varadarajan U. (2025). How “power couples” can help the United States win the global AI race. RMI, Rocky Mountain Institute, rmi.org, 20 febbraio.
  14. EOG (2024), OpenAI Invests in Nuclear Fusion. Energy Oil & Gas Magazine, energy-oilgas.com/news/openai-invests-in-nuclear-fusion/, 5 giugno.
  15. Deegan G. (2024). Google plan for new data centre rejected in face of electricity supply pinch. The Irish Time, irishtimes.com, 27 agosto.
  16. Dhiman R., Miteff S., Wang Y., Ma S.-C., Amirikas R., Fabian B. (2024). Artificial intelligence and sustainability - A review. Analytics 3(1): 140-164. DOI: 10.3390/analytics3010008.
  17. Gardner T. (2024). US closes $1.52 billion loan to resurrect Michigan nuclear plant. Reuters, reuters.com, 1 ottobre.
  18. Google (2024). Environmental Report, disponibile al sito: gstatic.com/gumdrop/sustainability/google-2024-environmental-report.pdf (consultato il 12 dicembre 2025).
  19. Goldman Sachs (2024). AI is poised to drive 160% increase in data center power demand. Goldman Sachs. Testo disponibile al sito: goldmansachs.com/insights/articles/AIpoised-to-drive-160-increase-in-power-demand (consultato il 14 maggio 2025).
  20. Gopal K. (2025). Trump signs executive orders to boost nuclear energy, reduce oversight. Inside Climate News, insideclimatenews.org, 23 maggio.
  21. Halper E. (2024). How Google and Meta data centers are keeping coal alive in Omaha. The Washington Post, washingtonpost.com, 12 ottobre.
  22. Halper E., O’Donovan C. (2024). AI is exhausting the power grid. Tech firms are seeking a miracle solution. The Washington Post, washingtonpost.com, 21 giugno.
  23. Harris D. (2024). Sustainable strides: How AI and accelerated computing are driving energy efficiency. Nvidia. Testo disponibile al sito: blogs.nvidia.com/blog/acceleratedai-energy-efficiency/ (consultato il 22 luglio).
  24. IEA (2024a). World Energy Outlook 2024. Ginevra: International Energy Agency.
  25. IEA (2024b). Electricity 2024. Analysis and Forecast to 2026. Ginevra: International Energy Agency.
  26. IEA (2024c). Renewables 2024. Analysis and Forecast to 2030. Ginevra: International Energy Agency.
  27. IEA (2025). Energy and AI. World Energy Outlook Special Report. Parigi: International Energy Agency.
  28. Kamiya G., Coroamă V.C. (2025). Data Centre Energy Use: Critical Review of Models and Results. EDNA-IEA 4E TCP.
  29. Kimball S. (2024). Microsoft’s carbon emissions up 30% since 2020 due to data center expansion. CNBC, cnbc.com, 15 maggio.
  30. KPMG (2024). Data Centers in Ireland. KPMG, kpmg.com, 21 novembre.
  31. Leon L.F.A. (2024). AI and global climate change: The political economy of data and energy in geographic perspective. Geography and Environment, 11(1). DOI: 10.1002/geo2.134.
  32. Li P., Yang J., Islam M.A., Ren S. (2025). Making AI less “thirsty”: Uncovering and addressing the secret water footprint of AI models. ArXiv: 2304.03271v4. DOI: 10.48550/arXiv.2304.03271.
  33. Li W., Li J.-P., Wang Y.-F., Stan S.-E. (2025). Is artificial intelligence an impediment or an impetus to renewable energy investment? Evidence from China. Energy Economics, 147. DOI: 10.1016/j.eneco.2025.108550
  34. Masley A. (2025). Using ChatGPT is not bad for the environment. The Weird Turn Pro, andymasley.substack.com, 13 gennaio.
  35. Mauludin M.S., Khairudin M., Asnawi R., Mustafa W.A., Fauziah T.S. (2025).
  36. The advancement of artificial intelligence’s application in hybrid solar and wind power plant optimization: A study of the literature. Journal of Advanced Research in Applied Sciences and Engineering Technology, 50(2): 279-293. DOI: 10.37934/araset.50.2.279293.
  37. McKinsey & Company (2024). Global Energy Perspective 2024. Rapporto.
  38. McLean J. (2024). Responsibility, care and repair in/of AI: Extinction threats and more-than-real worlds. Environment and Planning F, 3(1-2): 3-16. DOI: 10.1177/26349825241228586
  39. Meta (2023). Environmental Data Index. Meta Sustainability Report.
  40. Metz R. (2024). Google’s emissions shot-up 48% over five years due to AI. Bloomberg, bloomberg.com, 2 luglio.
  41. Microsoft (2024). How can we advance sustainability? Environmental Sustainability Report.
  42. Nost E. (2024). Governing AI, governing climate change? Geography and Environment, 11(1). DOI: 10.1002/geo2.138.
  43. NVIDIA (2024). Investor-Presentation, October 2024.
  44. Olatunde T.M., Okwandu A.C., Akande D.O., Sikhakhane Z.Q. (2024). Reviewing the role of artificial intelligence in energy efficiency optimization. Engineering Science & Technology Journal, 5(4): 1243-1256. DOI: 10.51594/estj/v5i4.1015.
  45. Parlamento Europeo (2023). Direttiva (UE) 2023/1791 del Parlamento europeo e del Consiglio del 13 settembre 2023.
  46. Penn I., Conger K. (2024). Why a Memphis community is fighting Elon Musk’s supercomputer. The New York Times, nytimes.com, 16 ottobre.
  47. Penn I., Weise K. (2024). Hungry for energy, Amazon, Google and Microsoft turn to nuclear power. The New York Times, nytimes.com, 31 ottobre.
  48. Pimenow S., Pimenowa O., Prus P. (2024). Challenges of artificial intelligence development in the context of energy consumption and impact on climate change. Energies, 17(23). DOI: 10.3390/en17235965.
  49. Sandalow D. (2024). Artificial Intelligence for Climate Change Mitigation Roadmap, seconda edizione, ICEF.
  50. Sætra H.S. (2021). AI in context and the sustainable development goals: Factoring in the unsustainability of the sociotechnical system. Sustainability, 13(4). DOI: 10.3390/su13041738.
  51. SELC (2025a). Musk’s xAI explores another massive methane gas turbine installation at second South Memphis data center. Southern Environmental Law Center, disponibile al sito: www.selc.org/press-release/musks-xai-explores-another-massive-methane-gasturbine-
  52. installation-at-second-south-memphis-data-center/ (consultato l’8 maggio 2025).
  53. SELC (2025b). Elon Musk’s xAI facility is using gas turbines in South Memphis, we’re taking action, Southern Environmental Law Center, disponibile al sito: selc.org/news/resistance-against-elon-musks-xai-facility-in-south-memphis-gets-stronger/ (consultato il 17 giugno 2025).
  54. Shehabi A., Smith S.J., Hubbard A., Newkirk A., Lei N., Siddik M.A.B., Holecek B., Koomey J., Masanet E., Sartor D. (2024). 2024 United States Data Center Energy Usage Report. Berkeley (California): Lawrence Berkeley National Laboratory.
  55. Spencer T., Singh S. (2024). What the data centre and AI boom could mean for the energy sector. International Energy Agency, disponibile al sito: iea.org/commentaries/what-the-data-centre-and-ai-boom-could-mean-for-the-energy-sector (consultato il 18 ottobre 2025).
  56. Srivathsan B., Sorel M., Sachdeva P. et al. (2024). AI power - Expanding data center capacity to meet growing demand, McKinsey & Company, mckinsey.com.
  57. Stein C. (2025). Trump signs executive orders to spur US ‘nuclear energy renaissance’. The Guardian, theguardian.com, 23 maggio.
  58. Strubel E., Ganesh A., McCallum A. (2019). Energy and policy considerations for deep learning in NPL. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Firenze: Association for Computational Linguistics: 3645-3650.
  59. Verdecchia R., Sallou J., Cruz L. (2023). A systematic review of green AI. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 13(4). DOI: 10.1002/widm.1507.
  60. Wang B., Wang J., Dong K., Nepal R. (2024). How does artificial intelligence affect high-quality energy development? Achieving a clean energy transition society. Energy Policy, 186. DOI: 10.1016/j.enpol.2024.114010.
  61. Wang Q., Li Y., Li R. (2024). Ecological footprints, carbon emissions, and energy transitions: The impact of artificial intelligence (AI). Humanities and Social Sciences Communications, 11.
  62. Wittenberg A. (2025). ‘How come I can’t breathe?’: Musk’s data company draws a backlash in Memphis. Politico, politico.com, 5 giugno.
  63. You J. (2025). How much energy does ChatGPT use? Epoch AI, epoch.ai, 7 febbraio.
  64. Zook M. (2005). The Geography of the Internet Industry. Venture Capital, Dot-Coms, and Local Knowledge. Malden, MA (USA): Blackwell Publishing.
  65. Zook M., McCanless M. (2022). Mapping the uneven geographies of digital phenomena: The case of blockchain. The Canadian Geographer/Le Géographe canadien, 66(1): 23-36.DOI: 10.1111/cag.12738.