Transformation of Agricultural Extension Communication in the Artificial Intelligence Era: A Systematic Literature Review

Abstract

The digital transformation of agriculture has reshaped how extension information is produced, delivered, and used by farmers. This study synthesizes research on agricultural extension communication in the artificial intelligence (AI) era, identifies major communication changes, and examines the benefits, challenges, and social implications of AI-based advisory services. A systematic literature review was conducted using the PRISMA 2020 framework. Articles published from 2019 to 2025 were searched in Scopus, ScienceDirect, SpringerLink, Taylor & Francis, and Google Scholar. After identification, duplicate removal, screening, eligibility assessment, and quality appraisal, 15 studies that directly met the inclusion criteria were analyzed through content analysis and thematic synthesis. The findings show that AI shifts extension communication from a conventional, linear, and top-down model toward a data-driven, interactive, personalized, and participatory advisory system. Machine learning, decision support systems, predictive analytics, agricultural chatbots, virtual assistants, and smart farming platforms improve information delivery, advisory personalization, and farmer decision-making. However, implementation is constrained by rural digital divides, limited technological literacy, data quality and bias, algorithmic opacity, inadequate infrastructure, and the risk of weakening farmer-extension worker relationships. The study proposes a Human-Centered AI Extension Communication Model that positions AI as an augmenting tool for participatory extension rather than as a substitute for the social, cultural, and facilitative roles of agricultural extension workers.