Abstract Taking specific instances of AI hallucinations encountered in the application of artificial intelligence within chemistry,such as the generation of incorrect reaction mechanisms, erroneous experimental results, or conceptual misunderstandings as a basis, this paper examines the underlying causes of AI hallucination and its potential risks in propagating scientifically inaccurate information in educational settings. To address these concerns, it is recommended that technical safeguards be implemented, including confidence thresholds and constraints that limit outputs to information derived from high-reliability, authoritative sources, thereby enhancing the accuracy and trustworthiness of generated content. Simultaneously, pedagogical guidelines should be established to regulate the use of AI in teaching and learning, while actively cultivating AI literacy among both educators and students.
ZHU Ma-Er, MO Fei, LIU Yong-Zhen. Challenges and Countermeasures of AI Hallucination in Chemistry Teaching: Risk Analysis and Educational Strategies Based on Empirical Cases[J]. Chinese Journal of Chemical Education, 2026, 47(9): 53-56.
