This chapter presents feedback and iterative refinement as core principles of communicative intelligence in human-AI interaction. It explores how conversational feedback loops transform static outputs into dynamic, co-developed exchanges. The chapter covers practical strategies such as checking understanding, requesting clarification, verifying outputs and interpretations, and ensuring mutual alignment between the user and the AI system. It addresses mechanisms for managing uncertainty, encouraging transparency, and refining AI-generated content through structured revision cycles. Emphasis is placed on feedback granularity, emotional tone, appreciation, and the importance of meta-feedback for evaluating the communication process itself. The chapter concludes with a discussion of systematic evaluation frameworks and long-term adaptation, highlighting how future AI systems can evolve communicatively by learning from feedback across modalities, contexts, and user preferences. Together, these principles establish feedback as a cornerstone of adaptive, transparent, and co-creative AI communication.