The integration of artificial intelligence (AI), particularly generative AI, is increasingly shaping science education by influencing teaching, learning, and assessment. While AI offers new opportunities for personalization and inquiry, it also raises pressing concerns about fairness, agency, and the erosion of educational values. To address these concerns, this chapter introduces the Responsible and Ethical Principles (REP) framework for the practice of AI, specifically in the context of science education. The REP framework distinguishes between responsible principles, which focus on how AI is implemented (e.g., fairness, privacy, human oversight), and ethical principles, which focus on why it is implemented (e.g., beneficial use, scientific integrity, respect for human values). This framework is essential for the practice of AI-supported science education. The chapter then situates the REP framework within a systems model of educational transformation across five areas: goals, procedures, materials, assessment, and outcomes. Scenario-based vignettes illustrate how the REP framework can guide classroom practice with AI supporting, rather than replacing, human judgment and reflection. The chapter closes with position statements that explain the necessity of the framework and its significance for the field. Together, the REP framework and the educational transformation model provide a practical lens to guide meaningful AI-supported science education in ways that remain responsive to context and values.