Improving farm productivity and profitability through AI diffusion: evidence from Punjab, Pakistan

Agriculture remains central to Pakistan's economy, with Punjab accounting for the largest share of national crop production. Despite the rapid global advancement of artificial intelligence (AI)-enabled agricultural technologies, their diffusion across farms in developing economies remains uneven and underexplored. This study examines the extent of AI diffusion among farmers in Punjab, Pakistan, and assesses its associations with perceived farm productivity and profitability.

Drawing on Diffusion of Innovations (DoI) theory, complemented by the Resource-Based View (RBV) and the Technology-Organisation-Environment (TOE) framework, the study adopts a quantitative, cross-sectional design using survey data from 101 farmers.

The findings indicate that while awareness and perceived benefits of AI are relatively high, actual adoption remains moderate (AI Diffusion: M = 18.27, SD = 3.71), indicating a gap between perceived value and implementation. AI diffusion was positively correlated with perceived farm productivity (r = 0.617, p < 0.001) and perceived farm profitability (r = 0.607, p < 0.001). The findings therefore demonstrate statistically significant associations between greater AI diffusion and stronger perceived farm performance, while not establishing causal effects.

The observed perception-adoption gap is consistent with structural barriers identified in previous research, including high investment costs, limited digital infrastructure and technical capability constraints. The study contributes context-specific evidence on digital agriculture in a developing economy by jointly examining AI diffusion, productivity and profitability. The findings highlight the importance of infrastructure, institutional support and capability development in facilitating wider and more effective adoption of AI-enabled agricultural technologies.