Environmental impact and net-zero pathways for sustainable artificial intelligence servers in the USA
Bibliographic record
Abstract
The rapidly increasing demand for generative artificial intelligence (AI) models requires extensive server installation with sustainability implications in terms of the compound energy–water–climate impacts. Here we show that the deployment of AI servers across the United States could generate an annual water footprint ranging from 731 to 1,125 million m3 and additional annual carbon emissions from 24 to 44 Mt CO2-equivalent between 2024 and 2030, depending on the scale of expansion. Other factors, such as industry efficiency initiatives, grid decarbonization rates and the spatial distribution of server locations within the United States, drive deep uncertainties in the estimated water and carbon footprints. We show that the AI server industry is unlikely to meet its net-zero aspirations by 2030 without substantial reliance on highly uncertain carbon offset and water restoration mechanisms. Although best practices may reduce emissions and water footprints by up to 73% and 86%, respectively, their effectiveness is constrained by current energy infrastructure limitations. These findings underscore the urgency of accelerating the energy transition and point to the need for AI companies to harness the clean energy potential of Midwestern states. Coordinating efforts of private actors and regulatory interventions would ensure the competitive and sustainable development of the AI sector. The rapid expansion of AI server installations in the United States poses sustainability challenges in terms of water usage and carbon emissions. A study now quantifies these potential impacts and outlines coordinated mitigation strategies for the AI sector to achieve net-zero.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".