The growth of generative artificial intelligence as a factor of investment transformation of the energy system based on the development of an integral index of energy intensity of data centers
Subject. The impact of the growth of generative artificial intelligence on investment processes in energy infrastructure. Objectives. Development of an integrated energy intensity index for data centers that aggregates the multidimensional pressure of digital infrastructure on the energy system into a single measurable indicator suitable for investment planning in the electric power industry. Methods. Pearson correlation analysis methods were used to evaluate the relationships between twelve pairs of data center load indicators and energy system investment metrics. To build the index, the method of weighted aggregation of three indicators was applied: load, dynamics and quality of infrastructure. The empirical basis was data on the US energy system from 2018 to 2024. Results. A high positive correlation has been established between the increasing load of data centers and investments in the energy system (Pearson coefficients from 0.79 to 0.97). An integral index has been developed, the value of which has increased from 1 to 1.82 over the study period, indicating an increase in pressure on the energy system by more than 80% relative to the baseline level. It has been confirmed that the average annual growth rate of data center demand for capacity is 2.6 times higher than the rate of commissioning of new generating capacities. Conclusions. The proposed index is a tool for quantifying the pressure of digital infrastructure on the energy system and can be used by regulators, investors and electric grid operators when making investment decisions in the electric power industry.
Keywords: generative artificial intelligence, data centers, energy intensity, investments in the electric power industry, integral index
References:
Masanet E., Shehabi A., Lei N. et al. Recalibrating Global Data Center Energy-Use Estimates. Science, 2020, vol. 367, no. 6481, pp. 984–986. DOI: 10.1126/science.aba3758
Mytton D., Ashtine M. Sources of Data Center Energy Estimates: A Comprehensive Review. Joule, 2022, vol. 6, no. 9, pp. 2032–2056. DOI: 10.1016/j.joule.2022.07.011
Chen Xin, Wang Xiaoyang, Colacelli A. et al. Electricity Demand and Grid Impacts of AI Data Centers: Challenges and Prospects. arXiv, 2025. DOI: 10.1109/TPWRS.2025.3550721
Huang J., Gopal S. Green AI – A Multidisciplinary Approach to Sustainability. Environmental Science and Ecotechnology, 2025, vol. 25, no. 100536. DOI: 10.1016/j.ese.2025.100536
Popad'ko N.V., Penar E.I. [Digitalization as a driver of energy consumption: the impact of data centers on global energy systems]. Ekonomika stroitel'stva, 2026, no. 3, pp. 50–53. (In Russ.) EDN: FCWUXQ
Veselov F.V., Solyanik A.I. [Midterm development conditions for the electric power industry of Russia under hard price constraints]. Problemy prognozirovaniya, 2020, no. 1, pp. 88–98. (In Russ.) EDN: PACZUM
Turing A.M. Computing Machinery and Intelligence. Mind, 1950, vol. 59, no. 236, pp. 433–460. Link
Vaswani A., Shazeer N., Parmar N. et al. Attention Is All You Need. arXiv, 2017. DOI: 10.48550/arXiv.1706.03762
Goodfellow I., Pouget-Abadie J., Mirza M. et al. Generative Adversarial Nets. arXiv, 2014. DOI: 10.48550/arXiv.1406.2661
Li Pengfei, Yang Jianyi, Islam M.A., Ren Shaolei. Making AI Less Thirsty: Uncovering and Addressing the Secret Water Footprint of AI Models. arXiv, 2024. DOI:10.48550/arXiv.2304.03271
Avgerinou M., Bertoldi P., Castellazzi L. Trends in Data Centre Energy Consumption under the European Code of Conduct for Data Centre Energy Efficiency. Energies, 2017, vol. 10, no. 10. DOI: 10.3390/en10101470
De Vries A. The Growing Energy Footprint of Artificial Intelligence. Joule, 2023, vol. 7, no. 10, pp. 2191–2194. DOI: 10.1016/j.joule.2023.09.004
Siddik M.A.B., Shehabi A., Marston L. The Environmental Footprint of Data Centers in the United States. Environmental Research Letters, 2021, vol. 16, no. 6. DOI: 10.1088/1748-9326/abfba1
Khudzhasaidov D.Kh., Rusina A.G., Matrenin P.V. et al. [Development of models for the forecasting of electricity consumption based on time series in isolated electrical power systems]. Elektrotekhnicheskie sistemy i kompleksy, 2020, no. 3, pp. 23–27. (In Russ.) DOI: 10.18503/2311-8318-2020-3(48)-23-27 EDN: YQJSSA
