Make AI a thinking partner, not just a machine that produces answers
[
](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEh_JzkGe2QDFuWnZR4DT9rU2UuMkrhpauhFp-RnozEdE0YYbx3ZnP9K7JQqb5UZ6lVm9RgB0DbwQaBvCQYy 51Oil516euxPd3E5fpe6ZC92pFIiBafNfWHgP88fJjQiOjvCwMRAosF3VRHrGCz7W1MZuyVK7 FeJ298UdPFVMr9Weo_9GFcqcNCYnmE7I4nu/s1536/%E7%A9%BF%E5%B1%B1%E7%94%B2.png)
This morning, I saw Teacher Zhang Junhong’s Message in the Facebook group “I love writing notes”. These words left a deep impression on me. He said:
Do simple things by yourself, come up with a framework and context (Context) for complex things and let AI do it, so that everything will become simpler. 
Yes! Over the past two years, I believe everyone has seen AI’s capabilities, but the question is how should we ask questions?
In the AI era, questions are more important than answers
In this era where AI models are readily available, we don’t actually lack answers. As we all know, the real challenge is can we ask deep enough, insightful enough questions?
When many people use ChatGPT, Claude or Gemini, they are always surprised by the fluent text and rich knowledge produced by these large language models, but at the same time they are confused: Why can AI always produce strategies with unique insights and styles when asked by others, but what I ask is mediocre, like a tasteless running account?
Well, the key is the density of questions.
This is not an abstract concept, but a combination of information density and thinking structure.
In traditional education, we are used to looking for the right answer; but in the AI era, the question itself is a thinking framework. Asking questions is not just about asking for information, but designing an invitation to think—inviting AI to push the boundaries of the unknown with you.
Question density: The amount of information determines the depth of thinking of AI
Let’s start with the picture below to see the difference in the amount of information caused by the density of questions.
[
Low-density questions are actually the most common in daily life. For example, many professionals like to ask: “How to do marketing?” Such questions are open in semantics, but poor in information. The reason is simple, because it lacks context, boundaries, and goals, AI can only give textbook-style, empty responses—it seems to be smooth and applicable to everyone, but it is therefore useless to everyone.
In contrast, high-density questions contain multi-dimensional constraints and goals. For example, you can ask: “Please help me design three versions of marketing copy for an online course that focuses on career transition groups. It is limited to 1,200 words. The tone should be both professional and motivational, and finally include a heart-warming call to action.”
In such a problem, AI already has enough decision-making context:
- Target audience (professional transfer group)
- Product attributes (online course)
- Output type (copywriting)
- Scope limit (within 1200 words)
- Emotional tone (professional + motivational)
When the information density of questions increases from 10% to 100%, the output of AI will naturally gradually move from generalization to specificity, from text generation to strategic reasoning.
In other words, the quality of questions determines the depth of AI’s thinking. Having said that, this is where many people misunderstand AI. AI 之所以厉害，不是因为大型语言模型愈强，它就愈懂你！ It’s because the more precise questions you can provide, the better it can perform!
AI is not an answer machine, but a thinking partner
In the picture below, we can see a simple but profound interaction model, that is, the relationship between the user and AI.
](It is true that this is not a one-way input-output relationship, but a co-evolution.
Users are responsible for asking precise questions, and AI feeds back in-depth insights; the real value of this process is not to obtain answers, but to expand the boundaries of our thinking.
I often tell many students that AI is actually not another Google. It is not an answer machine, but a thinking mirror. It allows you to see the structure of your thinking:
- When you ask vaguely, it will answer vaguely.
- When you ask specific questions, it can generate specific action plans.
- When you can ask layer by layer, it can even lead you to the source of insight.
The wisdom of AI does not lie in how much it knows, but in its ability to become a thinking companion, practicing thinking with you.
Having said that, every question you ask is actually training yourself: learning how to define problems more accurately, frame variables, and construct valuable conclusions.
The eight-round structure of in-depth dialogue: Let thinking spiral
Here, I would like to talk about the picture below in particular, which is the eight-round structure for creating in-depth dialogue.
[
This architecture is a thinking process diagram that I have summarized in recent years in teaching and AI application consulting. It applies not only to the interaction between people and AI, but also to in-depth conversations between people and even self-reflection.
Let’s look at these eight stages first:
- Open-ended exploration: Start with “What do I want to know?” This is the starting point for thinking.
- Focus on details: Begin to be specific, from vagueness to clarity.
- Ask for examples: Let abstract concepts have a visual basis.
- Comparative analysis: enter into systematic thinking and identify similarities, differences and cause and effect.
- Critical examination: challenge assumptions and examine biases.
- Situational application: Put knowledge back into the context of reality and test feasibility.
- Integration and reconstruction: Reorganize ideas and generate new models.
- Action transformation: finally turn insights into actions and complete the translation of knowledge.
These eight rounds of processes constitute a spiraling ladder of thinking.
Each round of questioning peels back new layers of knowledge. Real insights are often discovered not in the first round, but after the fifth round - when we start to reflect on premises, compare similarities and differences, and challenge assumptions, the “aha!” moment occurs.
From dialogue to co-creation: eight rounds of practical applications
Now, let’s use a concrete case to see how this architecture works.
Suppose you are a corporate training consultant and want to use AI to help you design a course called “AI Application in the Workplace.” If you just ask: “Please help me design an AI course for professionals,” AI will quickly provide a standard course outline without thinking, including: introduction, overview of generative AI, application cases, challenges and future trends, etc.
Yes, this is the first round of open exploration.
When you ask again: “If the class is for mid-level managers and the class is only three hours long, how should I adjust the content?”
At this time, you begin to enter the second and third rounds of questions, which are to specify the situation and provide examples of requirements.
Next, you may ask: “Please compare this course design with the AI application courses offered for college students.”
This is the fourth round of comparative analysis. At this time, AI will help you identify the differences in knowledge depth and practical orientation between the two.
When you further ask: “Please help me check whether this syllabus is too technology-oriented and ignores the decision-making aspect of managers?”
At this point, you have entered your fifth round of critical examination. At this point, AI may begin to challenge original designs and even propose alternatives.
Next, if you continue to ask: “Please help me simulate specific scenarios and students’ reactions when applied in the banking and financial industry.”
This is the sixth round of situational application, where you begin to verify the real-world applicability of your knowledge.
Then, you may ask AI: “Based on the above questions, please help me integrate an AI application curriculum suitable for financial industry executives.”
Well, this is the seventh round of integration and reconstruction. At this time, you are likely to find that AI will not only give you a set of mediocre answers, but will generate a new and exclusive knowledge structure.
Finally, when you ask: “Please help me design the promotional copy and enrollment strategy for this course.”
Congratulations, you have completed the eighth round of action transformation! At this point, AI is no longer just your think tank, but your strategy execution partner.
AI Learning Together: Retraining Thinking
The reason why such an eight-wheel architecture is important is not only because it makes AI smarter, but also because it allows us to learn to think again.
Based on my years of experience as a consultant in companies, I have found that many knowledge workers have gradually lost their patience for deep thinking in their long-term task-oriented work. There is no right or wrong in this, but we are too accustomed to looking for answers, too dependent on templates, and at the same time too afraid of falling behind others.
But the essence of AI dialogue is to constantly face unknowns and uncertainties.
Each round of questions is an exercise in thinking muscles. From fuzzy to clear, from input to reflection, and from single point to system, these attempts are all aimed at exercising the precious ability in the AI era, that is, cognitive enhancement.
Many people are accustomed to leaving their work to AI. Let’s not discuss whether it is appropriate to do so? You might as well think about a question first: What is the meaning and value of AI? I believe you can understand that the advent of AI is not to replace human thinking, but to help us restore and activate our thinking ability.
When you can have in-depth conversations with AI, you are actually relearning how to talk to the world, including: how to describe and dismantle complex problems, and how to generate new knowledge structures with the assistance of AI.
From questioner to designer: Design thinking for AI conversations
To truly make AI your thinking partner, we must change from mere users to designers willing to use our brains.
In the field of Design Thinking, there is an important principle called “Frame the problem”, which is to define the problem and create the solution. If you think about it, the same goes for the design of AI conversations.
To briefly summarize, a high-quality AI question should meet four conditions at the same time:
- Goal-oriented: What do you want AI to help achieve?
- Contextualized: Under what environment or constraints does the problem occur?
- Structured: In what form do you want AI to answer? (e.g. strategy steps, forms, templates)
- Sustainable questioning (Expandable): can extend the next round of dialogue.
Yes, these four points constitute the design grammar of AI dialogue. It makes every question not just an interaction, but a fun collaboration.
When you start talking to AI in this way, you will find that AI is not just a tool, but an engine that can think with itself. It can fill gaps in your knowledge and boldly challenge your assumptions. The most important thing is that it will make you think more three-dimensionally and consciously!
The birth of insight often begins in the fifth round
I often remind students: “Real insight begins in the fifth round.”
In the first four rounds, we are still in the information gathering stage; only in the fifth round - critical review, thinking really begins.
When entering this round, we must first stop and ask ourselves:
- Am I presupposing certain biases?
- What are the blind spots of this model? *Would the conclusion be different if viewed from another perspective?
The role of AI here is not to provide new knowledge, but to trigger reflection. It makes us realize that knowledge is not static but dynamically generated; insights are not given but discovered.
Upgrading thinking in the era of co-creation
It is true that we are entering a new era of learning. AI is not just an indexer of knowledge, but an amplifier of thinking.
If you want to continue to grow in this era, the key is not to learn how many tools, but to learn to collaborate and think with AI.
That means:
- Don’t simply outsource work to AI, but treat it as a partner.
- Don’t stop looking for answers, but continue to ask about the essence.
- Not content with having information, but refine your views and insights.
Every in-depth conversation is actually a cognitive upgrade. When you can combine questioning density, eight-round structure and collaborative learning, you will naturally have a new superpower, that is, let AI become your thinking accelerator.
Embrace AI and see a better version of yourself
The value of AI is not that it can replace human labor, but that it allows us to see a better version of ourselves. Just imagine, when you have in-depth conversations with AI again and again, you are actually having a conversation with your own thinking.
In that seemingly boring process, you will suddenly discover:
- Questions become more logical. *Thinking becomes more hierarchical.
- Become more confident in decision-making.
I know that many people are worried that jobs will be replaced by AI. In fact, it is not a threat to the future, but a mirror of the present. It reflects our curiosity, anxiety, curiosity and creativity.
Only those who are willing to keep asking, questioning and re-asking will not be ruthlessly eliminated by the times, and will also be the ones who can truly continue to evolve in the AI era.
Postscript
When we re-understand AI and no longer regard it as a simple tool, but an extension of thinking, you will find that every question is actually an opportunity for mental upgrade.
I hope we can all find a more sincere and profound version of ourselves in our conversations with AI, and also find the path to wisdom (https://copywriting.vista.tw/finding-strength-in-change/).
⋯ ✦ ⋯
🚀 If you are also looking for ways and opportunities to make full use of AI, welcome to learn more about courses and services.
📖 深入探索相關主題
