The Innovation Lifecycle of AI-Driven Agriculture: Causal Dynamics in University–Industry–Research Collaboration

Considering the current prediction from the Food and Agriculture Organization, food production needs an increase of over 70 percent by 2050, and the agriculture sector requires a boost that is also obtained by integrating novelty technologies and methods of the artificial intelligence (AI) field. This study investigates the innovation lifecycle of AI-driven agriculture through a causal inference analysis within the university–industry–research (UIR) collaboration framework. Using time series data from 1985 to 2023, collected from Web of Science (academic articles), CORDIS (research projects), WIPO (patents), and Crunchbase (start-ups), the study explores the causal dynamics among four key innovation pillars. Results from Granger causality and impulse-response analyses reveal a sequential innovation pathway in which start-up activity precedes research projects, followed by academic publications and patent filings. These findings highlight the interdependence of innovation stages and the evolving role of UIR collaboration in driving agricultural transformation. The study provides quantitative insights for policymakers, investors, and researchers to strategically foster innovation ecosystems in AI-driven agriculture.