Generative AI is accelerating change in schools. It is lowering the cost of information and knowledge while unsettling assessment, authorship, and trust. This paper reports on research that combines educator interviews, case studies from early adopters, and a review of recent scholarship to map the opportunities and the risks. The analysis identifies three consistent and relevant challenges: how to build confidence in safe and effective use, how to support educators and learners in adopting AI responsibly, and how to sustain motivation for change without losing sight of educational purpose. Findings show that compliance and policies by themselves do not address risks of bias or privacy, and that a broad range of stakeholders in schools are seeking both conceptual clarity and practical strategies. We propose a framework of Mindset, Mechanism, and Motivation as a way to guide responsible adoption. This model recognizes that AI can and should be responsibly integrated into school life as a form of support and guidance, but within the values and practices that shape teaching and learning.

‍