Enabling responsible AI-driven agri-food innovation in Ontario: A framework for analysis of adoption challenges and opportunities

AI adoption in the agri-food sector offers significant gains in productivity and competitiveness, but responsible implementation is essential to avoid stakeholder resistance and ethical concerns. This study examines the adoption of artificial intelligence (AI) technologies in Ontario’s horticultural and livestock sectors. Applying a systems perspective and responsible innovation, it identifies and categorizes emerging AI applications, develops a conceptual framework to capture technological, social, environmental, individual, and institutional factors, and proposes practical strategies to promote adoption. A structured literature review of peer-reviewed articles, government reports, and industry publications was conducted to manually classify AI technologies into content layer classifications: descriptive, diagnostic, predictive, and prescriptive, and map them to a framework. Diagnostic and prescriptive technologies dominate in horticulture, while AI applications in livestock are fewer and more evenly distributed across functional layers. Out of the 24 technologies identified, only four technologies, three in horticulture and one in livestock, demonstrated all analytical functions, highlighting the need for more integrated AI solutions. Key barriers include high cost, interoperability challenges, data privacy concerns, technical skill gaps, and limited digital infrastructure. Recommendations include promotion of targeted institutional support, operational efficiency, and ethical data governance. The framework provides practical guidance for responsible AI adoption and a foundation for future empirical research.