In recent years, precision agriculture has witnessed rapid integration of data-driven technologies to optimize pesticide use, minimize environmental harm, and improve crop health. Central to this transformation is the deployment of artificial intelligence (AI) models that predict optimal pesticide application strategies based on complex environmental, phenotypic, and agronomic data. However, the black-box nature of many AI systems poses a significant barrier to their widespread adoption by agronomists, farmers, and policymakers. This paper explores the application of Explainable Artificial Intelligence (XAI) in pesticide decision-making, focusing on how transparency and interpretability can bridge the trust gap in AI-powered crop protection solutions. We begin by outlining the current landscape of AI in agricultural pest management, highlighting the advancements in convolutional neural networks, decision trees, and ensemble methods for pest and disease detection. We then examine how XAI techniques—such as SHAP (SHapley Additive exPlanations), LIME (Local Interpretable Model-agnostic Explanations), and counterfactual reasoning—are being applied to explain model predictions in real-time pesticide advisory systems. The study emphasizes how interpretability enhances user confidence, facilitates regulatory compliance, and supports collaborative decision-making across the agricultural value chain. Further, we analyze several case studies where XAI frameworks improved the accuracy and acceptance of AI models in determining pesticide type, dosage, and application timing under diverse climatic and soil conditions. Challenges in integrating XAI into resource-limited farming systems, including computational overhead and model complexity, are discussed. Finally, the paper proposes a roadmap for embedding XAI into precision agriculture platforms to ensure sustainable, responsible, and human-centric AI deployment in pesticide management.