Mitigating risks, embracing potential: a framework for integrating generative artificial intelligence in geographical and environmental education

The rapid emergence of Generative Artificial Intelligence (GenAI) presents both opportunities and risks for geographical and environmental education, yet educators lack structured guidance for its implementation. This paper develops a theoretical framework for integrating GenAI into geography teaching and learning, synthesising insights from educational technology research, geographical education literature, and emerging AI applications. Through analysis of academic and professional sources, the study examines key GenAI capabilities relevant to geographical education, including personalised instruction through retrieval-augmented generation (RAG), interactive environmental simulations, and advanced spatial data visualisation. The proposed framework addresses critical challenges such as AI hallucinations, algorithmic bias, and spatial analysis limitations, while providing preliminary guidance for responsible AI adoption. This work establishes a foundation for future empirical research, with next steps including classroom-based testing and refinement of the framework across different educational contexts to validate its effectiveness in preserving core geographical thinking skills while leveraging AI capabilities.