Meeting growing energy demands sustainably remains a critical global challenge, particularly in rapidlydeveloping regions. This study presents an AI-driven machine learning model designed to optimize energy resourcedistribution in Kwara State, Nigeria. By analyzing key variables building type, settlement classification, city, and monthlyconsumption, the model accurately predicts energy demand patterns. Gradient Boosting emerged as the top-performingalgorithm, achieving a Mean Absolute Error (MAE) of 334.69 and Root Mean Squared Error (RMSE) of 390.12. Residualplots confirmed the model’s reliability, with errors randomly distributed around zero. The findings highlight AI’s potentialto enable data-driven energy management, offering policymakers actionable insights for sustainable planning.Keywords: Energy optimization, AI-based model, machine learning, Kwara State, sustainable development