Sustainable Computing for Digital Livestock: Reconciling Artificial Intelligence with Planetary Boundaries
Bibliographic record
Abstract
Artificial intelligence is transforming digital livestock farming, yet the same systems that improve welfare, efficiency, and emissions monitoring can impose large carbon costs from training, continuous inference, and hardware manufacture. This PRISMA-guided systematic review examines how Green AI can realign performance with environmental responsibility. We searched IEEE Xplore, Scopus, Web of Science, and the ACM Digital Library (January 2019–October 2025), screening 1,847 records and including 89 studies (61 with quantitative data). We address three questions: (RQ1) How do energy-efficient model designs reduce computational footprints while preserving accuracy? (RQ2) Which low-carbon machine-learning frameworks minimize training and inference emissions? (RQ3) How do sustainable infrastructures enable climate-positive deployments? Meta-analysis shows strong decoupling of performance from impact. Compression (pruning, quantization, distillation) achieves 70–95% parameter reductions with <5% accuracy loss. Lightweight architectures (e.g., MobileNet, EfficientNet) deliver 10–50× energy savings versus conventional CNNs, while neuromorphic systems achieve 200–1000× power reductions. Carbon-aware scheduling cuts emissions by ~70% via temporal and spatial workload placement; federated learning reduces communication energy by ~85% while preserving privacy; edge–fog–cloud hierarchies lower inference energy by ~87% by localizing computation. Six representative deployments report mean energy savings of 90.3% (85.9–99.96%) and cumulative CO₂ reductions of 2,175 kg with >91% accuracy retained. Key gaps remain: no ISO-aligned carbon metrics for agricultural AI; embodied emissions are rarely counted (17% of studies); accessibility for smallholders is limited; rebound effects are unquantified. We propose a roadmap prioritizing ISO-compliant accounting, low-cost solar or neuromorphic edge devices, rebound analysis, field validation, and multi-stakeholder Pareto optimization.
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 | 0.002 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.002 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| 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; both teacher heads agree on what is shown here.
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".