Artificial Intelligence and the Problem of Energy Consumption: Challenges and Opportunities for a Sustainable Future
Abstract views: 6
/ 
PDF downloads: 5
Keywords:
Artificial Intelligence, Energy Consumption, Environmental Sustainability, Green AI, Carbon EmissionsAbstract
In recent years, artificial intelligence (AI) systems have rapidly advanced, offering transformative
solutions that significantly enhance efficiency across various sectors such as healthcare, finance,
transportation, and education. However, the growing computational demands required to train and deploy
state-of-the-art AI models particularly those based on deep learning architectures—have raised serious
concerns regarding their environmental sustainability. Training large-scale models often involves the
consumption of vast amounts of electricity, resulting in substantial carbon emissions and a considerable
ecological footprint. This paper provides a comprehensive examination of the energy consumption
associated with AI systems, highlighting the underlying factors contributing to their environmental impact,
including model complexity, dataset scale, and infrastructure design. It also surveys recent efforts aimed at
mitigating these effects, including the development of energy-efficient algorithms, adoption of low-power
hardware architectures, and implementation of carbon-aware computing strategies. Furthermore, the
concept of “Green AI” is discussed as a paradigm shift towards sustainability-aware AI development,
advocating for the inclusion of energy and environmental metrics as core evaluation criteria in AI research.
The paper concludes by emphasizing the need for interdisciplinary collaboration and policy intervention to
align the progress of AI technologies with global sustainability goals.
Downloads
References
Gökgöz, B., Aydın, T., & Gül, F. (2024). Optimizing Memristor-Based Synaptic Devices for Enhanced Energy Efficiency and Accuracy in Neuromorphic Machine Learning. IEEE Access, 12, 154401–154417. https://doi.org/10.1109/ACCESS.2024.3482110
Han, S., Pool, J., Tran, J., & Dally, W. J. (2015). Learning both Weights and Connections for Efficient Neural Networks. http://arxiv.org/abs/1506.02626
Horowitz, M. (2014). 1.1 Computing’s energy problem (and what we can do about it). 2014 IEEE International Solid-State Circuits Conference Digest of Technical Papers (ISSCC), 10–14. https://doi.org/10.1109/ISSCC.2014.6757323
Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H. (2017). MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. http://arxiv.org/abs/1704.04861
Lacoste, A., Luccioni, A., Schmidt, V., & Dandres, T. (2019). Quantifying the Carbon Emissions of Machine Learning. https://doi.org/1910.09700
Patterson, D., Gonzalez, J., Le, Q., Liang, C., Munguia, L.-M., Rothchild, D., So, D., Texier, M., & Dean, J. (2021). Carbon Emissions and Large Neural Network Training. http://arxiv.org/abs/2104.10350
Roy, K., Jaiswal, A., & Panda, P. (2019). Towards spike-based machine intelligence with neuromorphic computing. Nature, 575(7784), 607–617. https://doi.org/10.1038/s41586-019-1677-2
Schwartz, R., Dodge, J., Smith, N. A., & Etzioni, O. (2019). Green AI. http://arxiv.org/abs/1907.10597
Schwartz, R., Dodge, J., Smith, N. A., & Etzioni, O. (2020). Green AI. Communications of the ACM, 63(12), 54–63. https://doi.org/10.1145/3381831
Strubell, E., Ganesh, A., & McCallum, A. (2019). Energy and Policy Considerations for Deep Learning in NLP. http://arxiv.org/abs/1906.02243
Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. http://arxiv.org/abs/1905.11946
