Artificial Intelligence and Agricultural Risk Management for Smallholder Cowpea Farmers and Processors in Niger State, Nigeria

Abstract— This study investigates the role of artificial intelligence (AI) in agricultural risk management among smallholder cowpea farmers and processors in Niger State, Nigeria. Using a mixed-methods approach and a sample of 200 respondents, the study assessed socio-economic characteristics, AI awareness and adoption patterns, perceptions of AI tool functionality, influencing factors, and adoption challenges. Results revealed that 62% of respondents were male, 43% aged between 31–45 years, and 47% had only primary or no formal education. The average farm size was 1.86 hectares, and 69% were cooperative members. Awareness of AI technologies was moderate to high, with 68% aware of AI-based weather forecasting, 62% aware of pest detection tools, and 54% familiar with price prediction platforms. However, only 42% had adopted any AI tool, and just 29% found them easy to use. Perception scores were highest for AI in weather forecasting (mean=2.91), pest detection (2.76), and risk mitigation (2.81), while ease of use (2.38) and device compatibility (2.44) were below the acceptance threshold. Regression analysis identified educational level, digital literacy, AI awareness, and extension contact as significant at the 1% level. Gender, farm size, and cooperative membership were significant at the 5% level, while age and access to credit were weakly significant (10%). Marital status, farming experience, and perceived risk level were not significant. Kendall’s Coefficient of Concordance (W=0.726, p < 0.001) revealed strong agreement on adoption challenges, with top-ranked constraints including low digital literacy (mean rank = 5.84), poor internet access (5.62), and high cost of digital tools (5.38).