| 英文摘要 |
With the rapid development of generative artificial intelligence (GAI), educators have gradually shifted from using a single AI tool to¬ward multi-AI collaboration. However, limited attention has been paid to how educators orchestrate multiple AI tools, verify information, and maintain professional agency across authentic, cross-task professional contexts. This study adopted an autoethnographic approach to analyze 16 months of the researcher’s professional practice, from April 2025 to July 2026, drawing on AI interaction records, work documents, and re¬flective notes. The findings indicate that multi-AI collaboration does not simply involve using more AI tools. Rather, educators select and con¬figure AI tools according to the nature of problems, task requirements, risks, and contextual conditions, while developing professional decisions through iterative human–AI interaction, cross-AI comparison, verifica¬tion against original sources, and contextualized interpretation. Based on these findings, this study proposes an“Educator Multi-AI Collaborative Work Model,”comprising seven interconnected and iterative processes. The findings demonstrate that effective AI use depends on educators’ability to orchestrate, verify, and integrate multiple AI resources while continuously maintaining professional judgment and authority over final decisions. |