Mapping recent trends in artificial intelligence research for sustainable agriculture: a bibliometric and systematic review

This work presents a bibliometric analysis and systematic literature review of artificial intelligence (AI) for sustainable agriculture (SA) using 645 documents in the Scopus database (2010–2025). The upward trend in AI publications is apparent in our analysis, which shows a yearly growth rate of 12.69%. Output is expected to rise in 2024. The largest contributors: India: 120 papers, China: 60 papers. AI research on sustainable agriculture has a strong interest in Machine learning (especially, deep learning), precision agriculture, and the IoT. They are applied for resource optimization, plant monitoring, climate change adaptation and the increase of efficiency. Nevertheless, there is an underrepresentation from Africa and South America in the literature, indicating a requirement for locally tailored AI solutions that account for different agroecological conditions. A major finding is an upward trend of interdisciplinary cooperation, and journals like Sustainability or Lecture Notes in Networks and Systems play a great role in this aspect. Yet, various challenges remain, including lack of data problems, lack of model generalizability difficulties, as well as ethical concerns like algorithmic bias and job displacement. Moreover, high costs of infrastructure and resistance from smallholder farmers are real challenges. Despite these naysayers, advances in AI have transformative promise. Solutions that combine systems, such as blockchain and CRISPR/CAS9, present routes to discoveries, but more research is required. Findings highlight the need for transparency, explainable AI frameworks, and strong international collaboration to advance smart technologies for sustainable agriculture, ensuring equitable, efficient, and context-specific innovation globally.