THE AI ENERGY DILEMMA: FINDING THE MIDDLE GROUND BETWEEN HIGH PERFORMANCE AND ECO-FRIENDLINESS
Article Sidebar
-
Green AI, Deep Learning, Energy Efficiency, Model Pruning, Sustainable Computing
Abstract
The exponential escalation of computational requirements for training and deploying Deep Learning models has precipitated an energy crisis, necessitating a critical reevaluation of the trade-off between algorithmic performance and environmental sustainability. This study aims to reconcile these conflicting demands by developing and validating a novel Dynamic Energy-Aware Pruning (DEAP) framework designed to maximize inference efficiency without compromising predictive accuracy. Employing a rigorous quantitative experimental design, we benchmarked state-of-the-art neural architectures, including ResNet-50 and Large Language Models (LLMs), across diverse hardware environments. The research utilized real-time telemetry to measure total energy consumption (Joules), thermal output, and carbon intensity () against standard accuracy metrics. Empirical results demonstrate that the proposed framework achieved a 42% reduction in energy consumption and stabilized hardware thermals, while maintaining predictive performance within a strict 1.5% non-inferiority margin compared to dense baselines. We definitively conclude that algorithmic sparsity effectively decouples high-level intelligence from excessive power usage, establishing a viable engineering paradigm for “Green AI” that aligns the trajectory of artificial intelligence with global decarbonization targets.
Full text article
References
Abdalla, A., Mohammed, M. M. A., Adedeji, O., Dotray, P., & Guo, W. (2025). Toward resource-efficient UAV systems: Deep learning model compression for onboard-ready weed detection in UAV imagery. Smart Agricultural Technology, 12, 101086. https://doi.org/10.1016/j.atech.2025.101086
Athanasoulias, S., Temenos, N., Kappos, I., Kokos, I., Navaro, P. A. G.-A., Ipiotis, N., Doulamis, A., & Doulamis, N. (2025). Isomorphic structured pruning of temporal CNNs for scalable NILM on edge devices. Energy Reports, 14, 3048–3061. https://doi.org/10.1016/j.egyr.2025.09.007
Bolón-Canedo, V., Morán-Fernández, L., Cancela, B., & Alonso-Betanzos, A. (2024). A review of green artificial intelligence: Towards a more sustainable future. Neurocomputing, 599, 128096. https://doi.org/10.1016/j.neucom.2024.128096
Cantini, R., Capalbo, M., & Talia, D. (2025). ZEP-NAS: Enabling green-aware model design via zero-cost emission proxy in neural architecture search. Array, 28, 100566. https://doi.org/10.1016/j.array.2025.100566
Che, W., Peng, C., & Zhuang, W. (2026). Green infrastructure for urban cooling and carbon neutrality: Building-integrated strategies from heat mitigation to public health. Urban Forestry & Urban Greening, 116, 129224. https://doi.org/10.1016/j.ufug.2025.129224
Chouksey, A., Rajan, A. K., Gurjar, V., Tiwari, R., & Mishra, P. K. (2026). The green paradox: The climate, environmental, and sustainability implications of artificial intelligence. Global Environmental Change Advances, 6, 100029. https://doi.org/10.1016/j.gecadv.2025.100029
Deb, N., & Rahman, T. (2025). An efficient VGG16-based deep learning model for automated potato pest detection. Smart Agricultural Technology, 12, 101409. https://doi.org/10.1016/j.atech.2025.101409
Dwivedi, P., & Kajal, M. (2025). Energy-aware and dynamic training of deep neural networks (EADTrain) for sustainable AI. Journal of Visual Communication and Image Representation, 112, 104582. https://doi.org/10.1016/j.jvcir.2025.104582
Fukase, V. Y., Gama, H., Bueno, B., Libanio, L., Reali Costa, A. H., & Jordao, A. (2025). One Period to Rule Them All: Identifying Critical Learning Periods in Deep Networks. International Neural Network Society Workshop on Deep Learning Innovations and Applications 2025, 264, 270–279. https://doi.org/10.1016/j.procs.2025.07.138
Gaurav, A., Pan, V. S.-H., Arya, V., Raman, R., Gupta, B. B., & Chui, K. T. (2026). AI-driven lightweight CNN model for sustainable vegetable classification in smart food systems. Green Technologies and Sustainability, 4(1), 100257. https://doi.org/10.1016/j.grets.2025.100257
Hussain, H., & Tamizharasan, P. S. (2026). Chapter 19—The role of optimization techniques in achieving sustainable artificial intelligence. In P. N. Mahalle, G. R. Shinde, N. N. Wasatkar, & P. R. Anerao (Eds.), Transforming Industries, Empowering Societies (pp. 213–224). Elsevier. https://doi.org/10.1016/B978-0-443-32878-7.00008-0
Islam, I., Sristy, S. S., Jahan, R., Tareq, Md. M. R., Hasan, M., & Uddin, Md. P. (2026). Advancing solar radiation prediction with explainable AI and ensemble learning techniques. Telematics and Informatics Reports, 21, 100292. https://doi.org/10.1016/j.teler.2026.100292
Jeon, Y.-J., Hong, S., Lee, T. S., Park, S. H., Song, G., Seo, M.-G., Lee, J., Lim, Y., An, J.-T., Lee, S., Jeong, H.-Y., Park, S. J., Lee, C., Jung, D.-H., & Kwon, C.-T. (2025). Volumetric Deep Learning-Based Precision Phenotyping of Gene-Edited Tomato for Vertical Farming. Plant Phenomics, 7(3), 100095. https://doi.org/10.1016/j.plaphe.2025.100095
Kamelian Rad, M., Neri, F., Moschoyiannis, S., & Bauer, R. (2026). Topographical sparse mapping: A neuro-inspired sparse training framework for deep learning models. Neurocomputing, 659, 131740. https://doi.org/10.1016/j.neucom.2025.131740
Mallick, M. T., Banerjee, S., Thakur, N., Saha, H. N., & Chakrabarti, A. (2025). Evaluation of State-of-the-Art Deep Learning Techniques for Plant Disease and Pest Detection. Computers, Materials and Continua, 85(1), 121–180. https://doi.org/10.32604/cmc.2025.065250
Mei, X., Zhu, T., Zhong, B., Wu, W.-M., Li, N., Wang, Y., Liu, X., Liu, R., Abdul, R., Yi, S., & He, Y. (2026). Artificial intelligence in microplastics domain: Current progress, challenges, and sustainable prospects. Journal of Hazardous Materials, 503, 141233. https://doi.org/10.1016/j.jhazmat.2026.141233
Naser, M. Z. (2026). When machine learning models retire, decay, or become obsolete: A review on algorithms, software, and hardware. Renewable and Sustainable Energy Reviews, 226, 116231. https://doi.org/10.1016/j.rser.2025.116231
Nuhash, M. I., Sohag, M., Ramit, S. S., & Tusher, R. T. H. (2025). A deep ensemble learning and explainable AI framework for accurate bottle gourd disease diagnosis and deployment. Smart Agricultural Technology, 12, 101541. https://doi.org/10.1016/j.atech.2025.101541
Parsoya, R., Bisht, B., Vlaskin, M. S., Jaiswal, K. K., Chauhan, P. K., Tripathi, M. K., Kurbatova, A., Rajput, V., & Kumar, V. (2026). AI-based wastewater treatment for a circular economy and sustainable management of PFAS, heavy metals, microplastics, and antibiotics. Cleaner Water, 5, 100189. https://doi.org/10.1016/j.clwat.2025.100189
Qadri, Y. A. (2026). Chapter 9—Role of artificial intelligence in supporting the development of Green IoT. In M. A. Jamshed & A. A. Shah (Eds.), Design and Analysis of Green and Sustainable IoT Technologies for Future Wireless Communications (pp. 197–214). Academic Press. https://doi.org/10.1016/B978-0-44-333000-1.00014-6
Rahaman, M., Southworth, J., Amanambu, A. C., Tefera, B. B., Alruzuq, A. R., Safaei, M., Hasan, M. M., & Smith, A. C. (2025). Combining deep learning and machine learning techniques to track air pollution in relation to vegetation cover utilizing remotely sensed data. Journal of Environmental Management, 376, 124323. https://doi.org/10.1016/j.jenvman.2025.124323
Sahu, M., Dash, R., Kumar Mishra, S., Humayun, M., Alfayad, M., & Assiri, M. (2024). A deep transfer learning model for green environment security analysis in smart city. Journal of King Saud University - Computer and Information Sciences, 36(1), 101921. https://doi.org/10.1016/j.jksuci.2024.101921
Shome, S., Das, Suman, Das, Saumya, & Pal, D. (2025). An extensive review of THz communication in 6G: Facilitating technologies with edge computing and native AI. Franklin Open, 13, 100434. https://doi.org/10.1016/j.fraope.2025.100434
Song, Z., Gu, Y., Liu, H., Zou, T., Lin, Y., & Ye, K. (2026). Application of deep learning in wind, solar, and ocean energy: An analysis of prediction, optimization, and operation & maintenance. Renewable and Sustainable Energy Reviews, 230, 116663. https://doi.org/10.1016/j.rser.2025.116663
Tmamna, J., Fourati, R., Ben Ayed, E., Passos, L. A., Papa, J. P., Ben Ayed, M., & Hussain, A. (2024). A binary particle swarm optimization-based pruning approach for environmentally sustainable and robust CNNs. Neurocomputing, 608, 128378. https://doi.org/10.1016/j.neucom.2024.128378
Wadhwa, D., & Malik, K. (2026). Deep learning generalized hybrid models for multi-species crop disease classification with explainable insights. Engineering Applications of Artificial Intelligence, 164, 113317. https://doi.org/10.1016/j.engappai.2025.113317
Waseem, M., Sajjad, M. M., Naqvi, L. H., Majeed, Y., Rehman, T. U., & Nadeem, T. (2025). Deep learning model for precise and rapid prediction of tomato maturity based on image recognition. Food Physics, 2, 100060. https://doi.org/10.1016/j.foodp.2025.100060
Xia, Y., Li, Y., & Liu, S. (2025). A LIME-LSTSNM approach based green building sustainability prediction using BIM design. Sustainable Computing: Informatics and Systems, 47, 101155. https://doi.org/10.1016/j.suscom.2025.101155
Yamsani, N., & Chenna Reddy, P. (2026). EdgeSched-DQN: An intelligent deep reinforcement learning-based framework for optimized task scheduling in edge-cloud environments. Array, 29, 100645. https://doi.org/10.1016/j.array.2025.100645
Zhu, J., Zhao, M., Zhang, T., & Li, R. (2025). The role of ESG score in forecasting China corporate green bond issuance: Evidence from tree-based learning models. Sustainable Futures, 10, 101324. https://doi.org/10.1016/j.sftr.2025.101324
Authors
Copyright (c) 2025 James Scott, Olivia Davis, Jessica Green
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.