Artificial intelligence technologies in agronomy: Current trends and future perspectives: A review

The global agricultural sector faces mounting pressures from climate variability, resource scarcity, and the imperative to feed a projected population of nearly ten billion by 2050. Conventional agronomic practices have proven increasingly insufficient in addressing the spatial and temporal heterogeneity inherent in agricultural systems. Artificial intelligence (AI) has emerged as a transformative paradigm, offering data-driven methodologies that span from predictive crop modelling to autonomous field management. This critical review synthesizes the current landscape of AI applications in agronomy, examining the scientific foundations of machine learning, deep learning, computer vision, and emerging generative architectures. The manuscript critically evaluates the deployment of Random Forest, CNN, YOLO, reinforcement learning, and IoT-integrated systems across seven major application domains, including yield prediction, precision nutrient management, irrigation scheduling, weed and pest surveillance, climate-smart agronomy, and decision support systems. The review specifically examines why high reported accuracies on benchmark datasets frequently fail to translate into reliable field performance, analyzing the technical, economic, and social barriers that constrain adoption. While significant advances have been demonstrated in controlled environments, persistent challenges including data scarcity, model generalization failures, high infrastructure costs, farmer trust deficits, and algorithmic opacity continue to impede widespread deployment. The review identifies six transformative research trends, from edge computing and explainable AI to digital twin architectures and federated learning and proposes a strategic roadmap for transitioning AI from experimental promise to agronomic practice. The findings underscore that realizing the full potential of AI in agronomy demands not merely algorithmic innovation but interdisciplinary collaboration, farmer-centric design, and policy frameworks that address data governance, ethical considerations, and equitable access across diverse agricultural systems.