Digital Intelligence–Agricultural Economy Integration and Agricultural Technological Choice Under Land Constraints: Evidence from China

Under increasingly binding cultivated land constraints, the effects of digital and intelligent technologies on agriculture depend not only on whether they improve productivity, but also on how they reshape the direction of technological progress among capital, labor, and land. Using provincial panel data from China, this study constructs an index of digital intelligence–agricultural economy integration (DIAI) and measures capital–land and labor–land technological bias based on a normalized factor-augmenting CES production function. The results show that DIAI significantly promotes capital-augmenting and labor-augmenting technological progress relative to land, indicating that digital intelligence reshapes agricultural production mainly by strengthening the technological capacity of non-land factors around a constrained land base. Mechanism analysis shows that the capital–land technological bias is affected through both the relative input scale and relative price channels, whereas the labor–land technological bias is mainly associated with changes in relative factor scale. Further analysis reveals that these effects are concentrated in regions with scarce cultivated land, while no significant effects are found in land-abundant regions. This suggests that land scarcity is an important condition shaping the technological choice effect of digital intelligence. Overall, DIAI does not reduce the importance of cultivated land; rather, it reorganizes the way capital and labor are combined with land and strengthens technological adjustment where land constraints are more binding. More importantly, the asymmetric mechanisms imply that policies for capital-oriented technological adjustment should address both the allocation of intelligent capital and the economic incentives governing its use, whereas labor-oriented adjustment should focus more on labor organization, skill upgrading, and the effective integration of workers with digital production systems. These findings provide new evidence on how digital intelligence influences agricultural technological choice and offer implications for improving factor allocation and promoting intensive and sustainable use of cultivated land.