EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Artificial Intelligence’s Hidden Footprint in Environmental Engineering: A Life-Cycle Risk Assessment and the AI-ERAF Governance Framework

F. A. Samiul Islam

American Journal of Data Science and Artificial Intelligence · 2025

DOI: 10.54536/ajdsai.v1i2.6293 ↗

Kurz gefasst

Fachartikel in der Zeitschrift American Journal of Data Science and Artificial Intelligence, 2025. Verfasst von F. A. Samiul Islam. Die Arbeit wurde bisher 1-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Artificial Intelligence is changing the game in environmental engineering. It’s great at things like advanced monitoring, making predictions, and squeezing more out of our resources. But here’s the catch: while AI promises to help the planet, it’s also piling on its own environmental problems. This study takes a hard look at the downside. This paper digs into just how much energy AI uses, how much greenhouse gas it lets off, the resources it eats up, and all the electronic junk it leaves behind from the moment someone trains a model to the day that hardware ends up in a landfill. The numbers are pretty wild. Training one big deep learning model can pump out hundreds of metric tons of CO2, mostly because data centers and supercomputers burn through so much electricity. GPUs and high-end chips need rare earth elements, which means more mining and more strain on the planet, especially in places where energy still comes from fossil fuels. This study took a life-cycle approach and pulled in ideas from global sustainability standards. One thing stood out: there’s almost no clear or open environmental reporting for AI tech. To make sense of all these impacts, this paper built the AI Environmental Risk Assessment Framework (AI-ERAF). It sorts out the pressures AI puts on the environment into three buckets: operational, systemic, and ethical. What’s clear is that if we let AI keep growing without guardrails, carbon emissions will keep climbing, electronic waste will pile up, and global inequalities will get worse. But there are ways to fix this. Powering data centers with renewables, designing more energy-efficient AI models, and using circular resource management driven by smart policies can really help. AI has the power to solve environmental challenges, but it also creates new risks we can’t ignore. If we want the benefits without the baggage, we need better rules, transparent audits, and AI that puts the planet first. This study lays the groundwork for responsible AI policies in environmental engineering, showing how we can close the gap between technological progress and true ecological sustainability.

Zitieren

APA

Islam, F. A. S. (2025). Artificial Intelligence’s Hidden Footprint in Environmental Engineering: A Life-Cycle Risk Assessment and the AI-ERAF Governance Framework. American Journal of Data Science and Artificial Intelligence. https://doi.org/10.54536/ajdsai.v1i2.6293

Vancouver

Islam FAS. Artificial Intelligence’s Hidden Footprint in Environmental Engineering: A Life-Cycle Risk Assessment and the AI-ERAF Governance Framework. American Journal of Data Science and Artificial Intelligence. 2025. doi:10.54536/ajdsai.v1i2.6293

BibTeX
@article{Islam2025,
  author = {Islam, F. A. Samiul},
  title = {Artificial Intelligence’s Hidden Footprint in Environmental Engineering: A Life-Cycle Risk Assessment and the AI-ERAF Governance Framework},
  journal = {American Journal of Data Science and Artificial Intelligence},
  year = {2025},
  doi = {10.54536/ajdsai.v1i2.6293}
}
Kostenlos starten →