Attipoe (2026) Artificial Intelligence Applications in Climate-Smart Agriculture: A Critical Review of Evidence, Implementation and Responsible Innovation

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Asian Journal of Advances in Agricultural Research
• Year: 2026
• Date: 2026-08-03
• Authors: Sonny Gad Attipoe
• DOI: 10.9734/ajaar/2026/v26i8746

Research Groups

[Not specified in the provided text]

Short Summary

This critical narrative review evaluates the application of artificial intelligence (AI) in climate-smart agriculture, concluding that while AI is effective for specific perception and prediction tasks, its actual contribution to climate-smart goals is often limited by poor model transferability and a lack of causal evidence.

Objective

• To evaluate how AI contributes to climate-smart agriculture, identify where evidence is strongest, and determine the methodological and institutional constraints that hinder reliable translation into practice.

Study Configuration

• Spatial Scale: Global (synthesis of literature across various regions and systems).
• Temporal Scale: 2010 to May 30, 2026.

Methodology and Data

• Models used: Critical narrative review and evidence synthesis.
• Data sources: Publicly accessible scholarly indexes, repositories, and verified DOI records.

Main Results

• High Confidence Areas: AI shows strong performance in well-bounded perception and prediction tasks with abundant labeled data, specifically image-based disease recognition, crop mapping, and certain yield-estimation applications.
• Technical Limitations: Model reliability weakens significantly when transferred across different seasons, regions, cultivars, or management systems.
• Evaluation Gap: Most studies rely on retrospective accuracy metrics and random data splits rather than establishing causal effects on water use, greenhouse gas emissions, profitability, or resilience.
• Methodological Preference: Hybrid approaches—combining process knowledge, spatially structured validation, and human oversight—are more defensible than "black-box" AI deployments.
• Conclusion: AI is not inherently "climate-smart"; its utility depends on data representativeness, agronomic validity, energy costs, and accountable governance.

Contributions

• The paper shifts the discourse from using technical accuracy as a proxy for impact to requiring outcome-based assessments that treat equity and ecological effects as core performance criteria.

Funding

[Not specified in the provided text]

Citation

@article{Attipoe2026Artificial,
  author = {Attipoe, Sonny Gad},
  title = {Artificial Intelligence Applications in Climate-Smart Agriculture: A Critical Review of Evidence, Implementation and Responsible Innovation},
  journal = {Asian Journal of Advances in Agricultural Research},
  year = {2026},
  doi = {10.9734/ajaar/2026/v26i8746},
  url = {https://doi.org/10.9734/ajaar/2026/v26i8746}
}