- Brief Communication
- Published:
E-waste challenges of generative artificial intelligence
- Peng Wang ORCID:orcid.org/0000-0001-7170-14941,2 na1,
- Ling-Yu Zhang1 na1,
- Asaf Tzachor ORCID:orcid.org/0000-0002-4032-49963,4 &
- …
- Wei-Qiang Chen ORCID:orcid.org/0000-0002-7686-23311,2
Nature Computational Sciencevolume 4, pages818–823 (2024)Cite this article
3368Accesses
10Citations
1027Altmetric
Abstract
Generative artificial intelligence (GAI) requires substantial computational resources for model training and inference, but the electronic-waste (e-waste) implications of GAI and its management strategies remain underexplored. Here we introduce a computational power-driven material flow analysis framework to quantify and explore ways of managing the e-waste generated by GAI, with a particular focus on large language models. Our findings indicate that this e-waste stream could increase, potentially reaching a total accumulation of 1.2–5.0 million tons during 2020–2030, under different future GAI development settings. This may be intensified in the context of geopolitical restrictions on semiconductor imports and the rapid server turnover for operational cost savings. Meanwhile, we show that the implementation of circular economy strategies along the GAI value chain could reduce e-waste generation by 16–86%. This underscores the importance of proactive e-waste management in the face of advancing GAI technologies.
This is a preview of subscription content,access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
9,800 Yen / 30 days
cancel any time
Subscribe to this journal
Receive 12 digital issues and online access to articles
￥11,900 per year
only ￥992 per issue
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
This paper analyzes existing and publicly available data. All data sources used in this research are referenced in the main text or in Supplementary Information17. Source data for Figs. 1b,c and 2b,c are available with this paper.
Code availability
The main code of our approach (as well as datasets to run the code) is available17.
References
Crawford, K. Generative AI’s environmental costs are soaring—and mostly secret.Nature626, 693 (2024).
Singhal, K. et al. Large language models encode clinical knowledge.Nature620, 172–180 (2023).
Grossmann, I. et al. AI and the transformation of social science research.Science380, 1108–1109 (2023).
Jia, Z. et al. The importance of resource awareness in artificial intelligence for healthcare.Nat. Mach. Intell.5, 687–698 (2023).
Lannelongue, L. et al. GREENER principles for environmentally sustainable computational science.Nat. Comput. Sci.3, 514–521 (2023).
Mytton, D. & Ashtine, M. Sources of data center energy estimates: a comprehensive review.Joule6, 2032–2056 (2022).
Masanet, E., Shehabi, A. & Koomey, J. Characteristics of low-carbon data centres.Nat. Clim. Change3, 627–630 (2013).
Computing 2030: Building a Fully Connected, Intelligent World (Huawei, 2021).
Baldé, C. P., et al.Global E-waste Monitor 2024 (ITU/UNITAR, 2024);https://ewastemonitor.info/the-global-e-waste-monitor-2024/
Parvez, S. M. et al. Health consequences of exposure to e-waste: an updated systematic review.Lancet Planet. Health5, e905–e920 (2021).
Data Centres and Data Transmission Networkshttps://www.iea.org/energy-system/buildings/data-centres-and-data-transmission-networks (IEA, 2023).
Ambrogio, S. et al. An analog-AI chip for energy-efficient speech recognition and transcription.Nature620, 768–775 (2023).
Wen, W. et al. Learning structured sparsity in deep neural networks. InProc. 30th International Conference on Neural Information Processing Systems 2082–2090 (Curran, 2016).
Jonathan, O. Who’s making chips for AI? Chinese manufacturers lag behind US tech giants.Naturehttps://doi.org/10.1038/d41586-024-01292-1 (2024).
Nuss, P. & Eckelman, M. J. Life cycle assessment of metals: a scientific synthesis.PLoS ONE9, e101298 (2014).
Walden, J., Angelika, S. & Maroye, M. Digital product passports as enabler of the circular economy.Chem. Ing. Tech.93, 1717–1727 (2021).
Johnly233. E-waste-Challenges-of-Generative-Artificial-Intelligence: revised version (V1.1).Zenodohttps://doi.org/10.5281/zenodo.13790035 (2024).
Acknowledgements
This research was financially supported by the National Natural Science Foundation of China (72274187 to P.W., 71961147003 to W.-Q.C.) and CAS IUE Research Program (IUE-JBGS-202202 to P.W.). We thank E. Masanet and our other colleagues for their contributions, which have improved this study.
Author information
These authors contributed equally: Peng Wang, Ling-Yu Zhang.
Authors and Affiliations
Key Lab of Urban Environment and Health, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen, China
Peng Wang, Ling-Yu Zhang & Wei-Qiang Chen
University of Chinese Academy of Sciences, Beijing, China
Peng Wang & Wei-Qiang Chen
School of Sustainability, Reichman University, Herzliya, Israel
Asaf Tzachor
Centre for the Study of Existential Risk, University of Cambridge, Cambridge, UK
Asaf Tzachor
- Peng Wang
You can also search for this author inPubMed Google Scholar
- Ling-Yu Zhang
You can also search for this author inPubMed Google Scholar
- Asaf Tzachor
You can also search for this author inPubMed Google Scholar
- Wei-Qiang Chen
You can also search for this author inPubMed Google Scholar
Contributions
P.W. and L.-Y.Z. designed the research; L.-Y.Z., P.W. and A.T. led the drafting of the manuscript. P.W., L.-Y.Z. and W.-Q.C. contributed to the methodology; L.-Y.Z., P.W. and A.T. interpreted the results. All authors contributed to the final writing of the article.
Corresponding authors
Correspondence toPeng Wang,Asaf Tzachor orWei-Qiang Chen.
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Computational Science thanks Loïc Lannelongue, Mengmeng Wang and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editor: Kaitlin McCardle, in collaboration with theNature Computational Science team.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Supplementary Information
Supplementary Discussion, Figs. 1–8, Tables 1–4 and equations 1–7.
Supplementary Data 1
Table of studied scenarios, giving a brief description and key parameter configurations for each scenario studied in the main text.
Source data
Source Data Fig. 1
Statistical source data.
Source Data Fig. 2
Statistical source data.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Wang, P., Zhang, LY., Tzachor, A.et al. E-waste challenges of generative artificial intelligence.Nat Comput Sci4, 818–823 (2024). https://doi.org/10.1038/s43588-024-00712-6
Received:
Accepted:
Published:
Issue Date:
Share this article
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
This article is cited by
Modeling the increase of electronic waste due to generative AI
- Loïc Lannelongue
Nature Computational Science (2024)
