Session 3: "How to Use Generative AI and Cautions: Introduction (English)"
     Audio commentary generated with Google NotebookLM:

Recording Date: 2025/11/4 (Generated by program)
Lecture Date: 2025/11/4 SSS Global Forum (Science Tokyo, Midorigaoka Building 6, Midorigaoka Hall)
Audience of the Lecture: Participants of SSS Global Forum
Recording Release: 2025/11/5
Access to the video: Unrestricted
Lecturer: Toshio Kamiya (Tokyo Institute of Science, Institute for Integrated Research, Element Strategy MDX Research Center)
Lecture Materials:
Slides
Textbook

Overview:
This tutorial explains effective ways to use generative AI and important cautions.
Rather than merely a tool for improving work efficiency, it is presented as a means to deepen thinking and broaden the scope of discussion. The following points are covered:

1. Prompt engineering as the key to eliciting high-quality responses
2. Risks of misinformation generated by AI (hallucination)
3. Cognitive debt caused by excessive reliance on AI
4. Prompt bias due to skewed questioning
5. Role-based prompting to embed expert perspectives into AI
6. Multi-perspective analysis by assigning multiple roles to AI
7. Using Markdown syntax to convey complex instructions accurately