Jeremy and Seth discuss the importance of maintaining AI data communities for training models, highlighting the need for continuous data updates in programming. They delve into the potential future of AI models being self-sufficient in learning new languages and packages, potentially impacting the necessity of human-supplied data.