問98
頻出度★★☆
AI におけるプロンプトエンジニアリングの説明として, 適切なものはどれか。
令和8年度 ITパスポート試験 問98
テクノロジ系基礎理論 › 情報に関する理論
This question's explanation and your answers so far.
解説
問 98
Incorrectア
Acquiring universal rules and knowledge from a large number of phenomena or data is a description of machine learning, which is different from devising prompts.
Incorrectイ
Indicating on the screen, using symbols or character strings, that input is possible describes the prompt (the input-prompt symbol) itself, which is a different concept from prompt engineering, which refers to devising ways to draw out an answer.
- ウ
正解 Prompt engineering is the technique of devising and optimizing the content of the questions or instructions (prompts) given to the AI, the background information provided, the specification of the output format, and so on, in order to draw the intended answer out of a generative AI. Therefore, choice (ウ) — which states devising the content of questions or instructions given to the AI, the information provided, the specification of the output format, and so on, in order to obtain the intended answer — is correct.
Incorrectエ
A computational model that imitates, on a computer, the network formed by nerve cells is a description of a neural network; it concerns the model structure of the AI, not prompt engineering.
ここがポイント
- Prompt engineering is the technique of devising the questions, instructions, background information, and output format given to a generative AI so as to more easily obtain the intended answer.
- Distinguish from similar terms: machine learning (acquiring rules from data), the prompt (the symbol display that prompts input), and a neural network (a computational model that imitates neural circuits).
用語