The metal manufacturing process primarily uses CNC machine tools for cutting, with most energy consumption occurring during machining. To optimize energy use, this study applies AI-based machine learning algorithms to develop a user-friendly interface for predicting CNC machine energy consumption. Various factors affect energy use, including toolpath, cutting parameters, spindle load, temperature rise, current variation, and tool wear. Existing monitoring systems focus mainly on current variation, often neglecting machining parameters. This study incorporates additional factors such as spindle speed, cutting speed, and feed rate while considering tool wear levels to enhance prediction accuracy. The goal is to assist manufacturers in reducing CNC machine energy consumption, lowering costs, and promoting sustainability. This study measures voltage, current, spindle and bearing temperature variations, tool wear, spindle speed, cutting speed, and feed rate during CNC operations to train an energy consumption prediction model. The Taguchi method identifies key parameters, and three supervised learning algorithms—Deep Neural Network (DNN), Long Short-Term Memory (LSTM), and Back Propagation Neural Network (BPNN)—are used to develop the model. Validation is conducted using CNC machines at the Smart Machinery and Intelligent Manufacturing Research Center of National Formosa University. In the future, this study aims to enable users to simply import the part drawing into the developed plugin, which will simulate the NC-CODE toolpath and machining parameters to predict energy consumption before machining. Further industrial testing will be also conducted to help manufacturers estimate energy use, select low-energy CNC machines, and improve machining efficiency, supporting sustainable manufacturing.