교수를 위한 ChatGPT 활용법 (2023년 10월)

English summary

This video, presented by a university faculty member on October 26, 2023, describes ways for university and school educators to use large language models and AI chatbots such as ChatGPT and Claude for course preparation, teaching design, assessment, assignment development, and analysis of student feedback.

Key points

• A prompt is a user's instruction to a language model. It can be organized into three parts: an instruction specifying the task, role, and objective; context providing background materials such as a syllabus draft or lecture notes; and guidelines defining format, level of detail, constraints, or prohibited elements.
• Suggested applications include drafting syllabi and time-allocated lectures, comparing complex concepts, explaining mechanisms step by step, using analogies and clinical examples, improving lecture materials, anticipating student questions, and generating subject-related engagement materials.
• For true or false and multiple-choice questions, prompts can specify answer placement, additional distractors, brief rationales, and exclusion of all-of-the-above or none-of-the-above options. The presentation also describes linking question activities to real-time tools such as Kahoot and Mentimeter; placing the correct answer in Option 1 is intended for instructor review before shuffling.

Audience

University and medical educators, instructors, and school educators interested in educational uses of AI chatbots

Scope and limitations

The source is a secondary summary of a presentation's topics, examples, methods, and limitations from October 2023. It explicitly notes that AI-detection tools are unreliable, chatbot outputs may vary even when the same prompt is repeated, and AI-generated suggestions or preliminary grading require human review. The examples and stated benefits are practical illustrations from the presentation rather than results of a systematic comparative study, and their generalizability and performance in actual educational settings were not established.

Research context

The educational use of large language models for course design, assessment, and feedback analysis has a neutral thematic connection to the use of educational data and AI methods within medical AI and data science.