The chapter explores how generative AI models like ChatGPT are being integrated into robots for improved reasoning abilities and task performance. It discusses the impact of deploying reasoning for robot movements directly on the robot or remotely on a server, highlighting advancements in techniques like diffusion policy, ACT, and deep reinforcement learning. The conversation also delves into the importance of verification in robotics, the concept of retrieval augmented methods, and the role of physics simulations and digital twins in optimizing robot functionalities.