This chapter investigates the application of Artificial Intelligence (AI) models to forecast energy consumption in buildings, particularly focusing on estimating potential energy savings resulting from renovation actions. Accurate estimation of energy savings is essential for the effective implementation of energy efficiency measures. To address this need, an ensemble model is proposed for precise baseline energy consumption estimation, utilizing tree-based algorithms such as Random Forest, XGBoost, and LightGBM. A key feature of the model is its emphasis on explainability, which provides transparency and insights into the primary factors influencing baseline energy consumption. Experimental evaluations conducted on a cluster of buildings in Latvia validate the accuracy of the model. The results demonstrate the superiority of the ensemble model over individual models designed for energy consumption estimation, thereby offering a reliable and interpretable approach for assessing energy savings in the building sector.