Xiaohang Che, Yueyang Liu
Generative artificial intelligence (AI) has entered undergraduate pharmacology learning through chatbots, writing assistants and large language model (LLM) interfaces. Students now use these tools to clarify mechanisms, compare drug classes, draft case answers, prepare questions and search for medication-related information. The same convenience creates a course-level problem: a response that is fluent and well organized may still contain an incorrect mechanism, an unsupported drug claim, a missing contraindication or a recommendation that does not fit the patient described in a case. This narrative review examines recent work on AI in medical and pharmacy education, with particular attention to pharmacology-relevant teaching tasks. It identifies problems that are especially important for pharmacology courses: hidden AI use, weak source checking of AI-generated drug claims, mechanism explanations that remain descriptive rather than causal, medication suggestions that ignore patient variables and assessment practices that evaluate the submitted answer rather than the reasoning behind it. In response, the article proposes a source-checking process for formative pharmacology coursework tasks in which AI use is permitted, including pre-class preparation, classroom case work, chapter-level tasks and short revision notes. The process asks students to define the task, check drug claims against defined sources, rebuild the pharmacological mechanism, test the answer against patient context and revise with feedback and justification. By embedding AI use within source checking, mechanism reconstruction and patient-context appraisal, this approach may help pharmacology teachers turn generated answers into structured material for learning, discussion and formative assessment.