Abstract
The rapid spread of ChatGPT has turned generative artificial intelligence into a practical issue for school teaching rather than a distant technological prospect. In geography education, this shift is especially important because geography requires students to interpret spatial patterns, compare scales, evaluate evidence, and justify claims about environmental and social processes. This review article synthesizes recent studies on ChatGPT, critical thinking, argumentation, and geography education to propose a methodology for classroom use aimed at developing students’ argumentation and analytical thinking. The reviewed literature suggests that ChatGPT can support learning performance, engagement, and some dimensions of higher-order thinking when it is used within structured, inquiry-based instruction. At the same time, the literature also points to clear risks, including hallucinated content, inaccurate local knowledge, weak source evaluation, and overreliance on fluent but analytically shallow responses. Based on this synthesis, the article proposes a six-stage methodology for geography lessons: problem framing, contextual prompting, evidence verification, multi-perspective comparison, argument construction with rebuttal, and reflective revision. It is argued that ChatGPT should not function as a substitute for disciplinary reasoning, but rather as a scaffold for critique, comparison, and evidence-based discussion in geography classrooms.