Paper Contents
Abstract
Smallholder farmers, who constitute the backbone of global food security, face persistent challenges in accessing timely, localized, and actionable agricultural information. This study investigates the development and deployment of Farmer.Chat, a scalable, AI-powered, voice-enabled agricultural chatbot designed to bridge this critical knowledge gap. The system leverages Generative AI, Natural Language Processing (NLP), and Multi-Layer Perceptron (MLP) neural networks, along with Retrieval-Augmented Generation (RAG), to process structured and unstructured agricultural datasets including soil profiles, climate records, and crop-specific databases. Farmer.Chat delivers real-time, personalized, multilingual, and context-aware recommendations on crop management, pest control, weather prediction, and market insights. A field deployment across Kenya, India, Ethiopia, and Nigeria engaged over 15,000 farmers, spanning more than 40 value chains, and addressed 300,000+ user queries in six languages through a voice assistant interface that ensures accessibility for low-literacy users. Analysis of adoption patterns and outcomes reveals improved crop yields, greater uptake of sustainable practices, and measurable reductions in input waste and operational costs. These findings suggest that AI-powered conversational agents can transform agricultural extension services, enhance decision-making, and advance equitable access to information in resource-constrained rural settings.
Copyright
Copyright © 2025 Amol More. This is an open access article distributed under the Creative Commons Attribution License.
