[交互逻辑] ChatGPT's level of understanding: what it knows, what you can ask, how to ask, and which questions can get answers.

[baike]ChatGPT's Level of Cognition[/baike]
ChatGPT knows almost all human knowledge. It can also program, translate, write, and reason; it can co-create and even call other software to help you complete tasks. According to AI expert Wu Jun, before ChatGPT emerged, computers could already help us solve about 40% of difficult problems. Now that ChatGPT has emerged, the problems solved have reached at least 60%; the extra 20% is the result of its self-learning reasoning and integrated analysis.
[baike]Cognitive Differences from Humans[/baike]
If we compare and analyze human cognition and ChatGPT's cognition and understand the cognitive differences, then based on these circumstances you will know where you need to ask, which questions can get effective answers, which cannot, and how to ask differently.

Figure 1 compares the cognitive differences between humans and ChatGPT and yields four different situations. Then what do you think ? represents? Keep this in mind as we look at the next figure.



Figure 2 gives the answer. The situation in which both we and ChatGPT know is called the open area. Next is “you don't know, ChatGPT knows”; this is the second category, called the inquiry area. We don't understand, and ChatGPT can teach us, so we are inquiring. The inquiry area has the largest area. The third situation is “you know, ChatGPT does not know”; this is called the private area. Data in the private area is precious and has relatively high value, and it is also what ChatGPT wants to obtain. We usually use private data as our own knowledge base but do not make it public. The last situation is “everyonedoes not know.” The third and fourth situations together make up the previously mentioned 40% of unsolved problems.



[baike]Characteristics of the Differences[/baike]
No matter which area it is, we will use it. According to different areas, our prompt questioning methods are different. Only after fully understanding ChatGPT's cognitive ability and communication logic can we easily get answers. We need to master the techniques involved. The moderator will list the characteristics of these four areas respectively; only after you understand them can you draw inferences from one instance.

1. Open Area
The open area is “you know and ChatGPT also knows.” When both sides know, it is easy to talk. Whatever you ask, it answers; you can clearly know whether its answer is right or good, and correct it. The open area is usually for verifying problems, or for using your knowledge to educate others. It is a process of tuning and verification.

2. Inquiry Area
“ChatGPT knows, you don't know.” This involves unknown or unfamiliar fields. You may have only a smattering of knowledge about this field, or you may be interested in it. At this time you have only a vague concept, you don't know the specific way to ask, and you don't know how to begin. If you ask poorly, you may go off track. Therefore, learning an unknown or unfamiliar field requires a set of learning methods, or at least a correct beginning.

I am a beginner, and I want to understand [xx行业]. What questions should I ask?
I am a beginner, and I want to understand [AI行业]. What questions should I ask?
This is a general prompt, but it is very useful. It will not make you go off track; instead, it will quickly get you to the point. For systematic teaching on the inquiry area, see: https://www.niuboyi.com/thread-650-1-1.html

3. Private Area
The private area is usually our confidential area, our private data. We do not show it to others easily because data is valuable. “Only you know, and others do not know.” This part can usually be sold for money. The use of this part of the content is usually as training data: give the data to a robot to learn, build an auto-reply robot, and use your data to give answers to others—of course, for a fee. Reference content: https://www.niuboyi.com/thread-631-1-1.html

4. Exploration Area
“You don't know, and ChatGPT doesn't know either.” Everyone does not know. For example, Sora, which was very popular recently: the content we can see at present is only some public test data from the official source, and this data has not yet been updated into the latest ChatGPT knowledge base. When we write articles, we often most need first-hand information. When everyone does not know, if we want to approach the answer, we must use its analysis and reasoning abilities. This is a process of mutual discussion and analysis; you need to first provide it with some information and then start the dialogue.

This is some of the only currently available [sora] data [提供的内容]. Let's discuss together whether there are similar products or technologies on the market and how they perform.
In which industries do you think [sora] will play a role, and which fields will it affect?
After [sora] is released, what information do you think different audiences, such as the general public, experts, and social elites, most want to know?
[baike]How to Verify That You Know[/baike]Many times, you think you know, but you actually do not know. This preconceived error is called blind conformity; it will lead to errors in your cognition and thus affect practical tasks. Therefore, when you learn a new technology, you should adopt a verification method to prove that you really know it. Verification method:

I think I already know a lot about [XXX], and I think I am already competent for work in this area. Please give me 10 questions to test me as an assessment of my workplace ability.
I think I already know a lot about [chatgpt提示词工程师], and I think I am already competent for work in this area. Please give me 10 questions to test me as an assessment of my workplace ability.
[baike]Summary of Efficient Questioning[/baike]Understanding these four situations is crucial. If beginners see this article first, their thinking will be much clearer. Many people who have played with AI for a while also usually have a feeling of only half-understanding; sometimes they feel they ask very well, and sometimes they cannot get anything clear out of it. In fact, this is where the problem lies. For problems in different situations, different questioning methods should be adopted. Only after understanding clearly can you achieve twice the result with half the effort.