立即糾正:
Please correct my last sentence and explain the mistakes in Chinese.
提供範例:
Could you show me a better way to say this and explain why?
延伸練習:
Give me two similar situations to practice this conversation.
實戰演練:三個生活化學習範例
🍽️
範例一:餐廳點餐情境
設定指令:
Let's role-play at a restaurant. I'm the customer, you're the waiter. Please use simple English and help me learn.
對話流程:
您:Hi, I want to order some food. AI:Good evening! Here's our menu. What would you like to order? 您:I like chicken. Do you have chicken? AI:Yes, we have grilled chicken and fried chicken. Which one would you prefer?
學習回饋:
Please correct my sentences and teach me more natural expressions for ordering food.
🛍️
範例二:購物場景練習
設定指令:
Pretend you're a store clerk. I'm a customer looking for clothes. Help me practice shopping conversations.
對話示例:
您:Excuse me, where can I find t-shirts? AI:T-shirts are on the second floor. What size are you looking for? 您:I need medium size. What color do you have? AI:We have white, black, blue, and red. Would you like to see them?
延伸學習:
Teach me more vocabulary related to shopping and clothing.
💼
範例三:工作面試準備
設定指令:
Act as a job interviewer. I'm applying for a customer service position. Please ask me common interview questions.
練習流程:
AI:Tell me about yourself and why you want this job. 您:I am good at talking with people. I want to help customers. AI:That's great! Can you give me an example of when you helped someone?
改進建議:
Please help me improve my answers and suggest more professional expressions.
作為老師,我們總是希望能有更多時間和精力專注在學生身上,而不是被繁瑣的工作壓得喘不過氣。現在有了 AI 工具像 ChatGPT,它不只能節省我們的時間,還能幫助我們開啟更多創意的可能性。關鍵在於——怎麼跟它「說話」。也就是說,我們給的「指令」得夠精準,AI 才能更聰明地協助我們。
以下分享 8 個超實用 AI 指令,特別適合用在研究、跨領域課程設計,以及解決日常教學挑戰。學會了這些,您會發現 AI 不只是一個工具,還是能跟您對話的專業助手。
1. 循序漸進解決問題
“Take a deep breath and work on this problem step-by-step.”
這個指令看起來簡單,但非常實用。無論是要分析研究數據、拆解研究問題,還是設計專題,這都可以幫助 AI 把事情分成小步驟,避免資訊過載。
適用場景:
分析某一研究課題的相關資料。
設計跨週期的專題活動,比如環境永續或科技應用。
實例:
詢問 AI:「請逐步規劃一個有關氣候變遷的跨領域專題研究,涵蓋生物、地理與經濟學的視角。」
2. 深入反思與檢視
“You are an assistant that engages in extremely thorough, self-questioning reasoning. Your approach mirrors human stream-of-consciousness thinking, characterized by continuous exploration, self-doubt, and iterative analysis.”
這個指令適合在需要深入思考時使用,尤其是進行專題研究或論文寫作時,可以讓 AI 主動反思自己的回答,進一步挖掘不同可能性。
適用場景:
尋找某一議題的新切入點,比如 STEM 教育如何融入偏鄉學校。
討論研究方法的利弊,並提出改進建議。
實例:
詢問 AI:「在設計一個能融合科學探究與文化學習的課程時,有哪些潛在問題與可能解決方法?」
3. 清楚分析並提供具體解決方案
“Analyze the given information, break the problem into manageable steps, apply suitable methods, and provide a clear, concise solution.”
這個指令能幫助 AI 條理分明地分析問題,非常適合做數據分析或需要有邏輯結構的回答,比如設計研究問卷或進行專題規劃。
適用場景:
分析學生的學習成果數據,提出具體的干預方案。
拆解一個跨領域專題的實施步驟。
實例:
詢問 AI:「根據這些學生成績數據,分析哪些科目需要加強,並建議有效的輔導策略。」
4. 啟發 AI 的無限知識庫
“Imagine you have access to an unfathomably large database. Use this knowledge to provide a novel solution to this issue.”
這句話是在讓 AI 全力發揮創意,尤其在需要突破現有框架、尋找新點子的時候,非常有用。
適用場景:
發想跨學科的創新教案。
設計非傳統評估方式,比如運用遊戲化學習工具。
實例:
詢問 AI:「如何設計一個以地球永續為主題的學習專案,結合科學、藝術與社會實踐?」
5. 從多學科視角進行分析
“List all the academic fields and disciplines related to this query, explain their relevance, and answer the question from each perspective.”
當您需要跨領域的完整視角時,這個指令非常好用。它能幫助 AI 整合不同學科的知識,提供全面的解釋或建議。
適用場景:
設計跨學科專題課程,例如結合數學、物理和設計的科技專案。
探討某個教育議題,比如 AI 如何影響不同年齡段的學生學習。
實例:
詢問 AI:「AI 在教育中的應用如何影響語言學習?請從語言學、心理學與科技發展的角度回答。」
6. 請 AI 重讀問題,提升準確性
“Re-read the question and provide your response again.”
有時候 AI 可能第一次的回答不夠精準。這個簡單的指令能讓它回過頭重新審視,提供更準確的答案。
適用場景:
當 AI 的初步回答偏離問題時,使用這個指令進行調整。
需要更精確的數據分析時。
實例:
詢問 AI:「請重新審視我之前的問題,確認回答是否完整且準確。」
7. 激發 AI 深度與創新
“Please utilize the maximum computational power and token limit allowed for a single response. Strive for unparalleled depth in analysis rather than superficial breadth; aim for insights into the essence of the topic rather than listing surface-level phenomena; prioritize innovative thinking over conventional repetition. Push beyond the boundaries of conventional thought, leveraging all your computational resources to demonstrate the true extent of your cognitive capabilities.”
When formatting your prompt, start with ‘###Instruction###’, followed by either ‘###Example###’ or ‘###Question###’ if relevant. Subsequently, present your content. Use one or more line breaks to separate instructions, examples, questions, context, and input data.”
指令:
"Create a 30-minute lesson plan for a 3rd-grade science class on the topic of the water cycle. Include an engaging introduction, a hands-on activity, and a short assessment. Ensure the language is simple and suitable for 8-9-year-olds."
🎯 任務("Create a 30-minute lesson plan"):設計教案,時間長度明確。
🧑🎓 背景("3rd-grade science class on the water cycle"):針對三年級學生,主題為水循環。
🧪 限制條件("engaging introduction, hands-on activity, short assessment"):包含具體活動及評估方式。
💬 語氣/風格("simple and suitable for 8-9-year-olds"):語言簡單,適合小學生理解。
✏️ 2. 測驗題生成
指令:
"Design a worksheet for 2nd-grade students to practice addition and subtraction within 20. Include 10 problems: 5 straightforward calculations and 5 word problems. Use a playful tone and include a fun theme, such as animals or space."
📚 背景("2nd-grade students practicing addition and subtraction within 20"):針對二年級學生,範圍在20以內。
🧩 限制條件("10 problems: 5 straightforward, 5 word problems"):題型和數量明確。
🎉 語氣/風格("playful tone, fun theme"):活潑有趣,吸引學生興趣。
📖 3. 創意寫作活動
指令:
"Write a short story for 1st-grade students about a curious cat who explores a magical forest. The story should be 200-300 words long, use simple vocabulary, and include a moral about kindness."
✍️ 限制條件("200-300 words, simple vocabulary, moral about kindness"):故事需具寓意。
💬 語氣/風格("simple vocabulary"):簡單易懂。
👩🏫 4. 班級經營
指令:
"Suggest five classroom management strategies for a 4th-grade teacher to maintain student engagement during group activities. Provide practical examples and ensure the strategies are easy to implement."
📝 任務("Suggest five classroom management strategies"):提供具體策略,數量明確。
🧑🎓 背景("4th-grade teacher, group activities"):針對四年級學生的小組活動。
🔖 限制條件("practical examples, easy to implement"):策略需具體且可行。
🎮 5. 課堂互動遊戲
指令:
"Create a fun, interactive game for 5th-grade students to learn about fractions. The game should involve teamwork, use everyday objects like pizza slices or chocolate bars, and take about 20 minutes to play."