Next Generation (NextG) wireless networks are critical for supporting advanced applications such as autonomous driving and smart factories. However, current communication systems are reactive and can only act on observed network information. This prevents them from anticipating the immediate needs of safety critical applications. This project addresses this limitation by developing a framework that allows the wireless network to become proactive rather than reactive. By integrating data from environmental sensors like cameras and lidars, the network can predict the location and future resource needs of a user before those needs are explicitly required. This capability will make autonomous transportation and smart factories safer and more efficient. To prepare the future workforce for these advanced wireless technologies, the project supports science and engineering education through the development of experiential robotics games for high school students. The research team will also engage the public through interactive museum demonstrations to promote widespread understanding of the next generation of wireless technologies. This project develops the Artificial Intelligence (AI) driven Integrated Sensing, Computing, and Communication framework to enable proactive network resource management. First, the project creates deep learning models using a Multi-Headed Transformer Architecture to translate raw sensor data into a predictive state representing the physical context and intent of the user. Second, the investigator will design a Hierarchical Graph Neural Network to model interactions among multiple users and resolve conflicting resource demands directly at the network edge. Third, the framework employs Multi-Agent Reinforcement Learning to map these user states into optimal resource allocation policies across different network layers. To ensure these complex AI models can operate on physical hardware, the research team will formulate a decentralized computing al