AI is not just an efficiency tool but a catalyst for cognitive upgrades. This article uses a series of practical examples to demonstrate how AI can be used to deeply analyze problems, stimulate insights, and continuously refine thinking through interactions.
Recently, I learned some ways to use AI to improve my cognition, and combined it with my own understanding to practice it myself.
AI’s capabilities are rapidly evolving, not only answering questions but also performing complex tasks.
For example, if you assign an AI a task, it will think deeply first, then call various tools, and finally output the corresponding results.
In the process of communicating with AI, we can use these prompt words: insight, insight, analysis, perspective, interpretation, deduction, deconstruction, traceability, speculation, fusion, and inspiration.
It can be difficult for humans to understand these words, but they can lead AI to better results because AI is already smarter than the vast majority of people.
AI is not only an assistant to people, but also a mentor to people.
With the help of AI, you can take your cognition to the level of an expert if you are willing to delve into it.
Communicating with advanced AI for a long time and getting world-class responses will change your brain, make you more sensitive to high-value information, and quickly grasp the essence of things.
After 10 years of interaction design, why did I transfer to product manager?
After the real job transfer, I found that many jobs were still beyond my imagination. The work of a product manager is indeed more complicated. Theoretically, the work of a product manager includes all aspects of the product, from market research, user research, data analysis…
View details >
We might as well ask AI questions frequently, such as:
- What deeper insights do you have on this?
- Is there a deeper insight than this?
- If I were a character in a movie, what do you think I would be?
- What advice do you have for me personally?
……
AI can generate countless results, but it is up to humans to choose which result.
AI can imitate countless styles, but which style is more beautiful requires human taste.
AI can perform countless tasks, but it takes human thinking about why it does it.
As AI becomes more powerful and efficient, the most “soft” things become the hardcore.
Whenever I see something good or have a new idea, as long as I want to know more, I send it to the AI and say three words: please analyze.
For example, I prefer to watch the science fiction drama “Black Mirror”, and after watching it, I feel unsatisfied, so I say to the AI: Please analyze “Black Mirror” Season 6, Episode 3.
AI will analyze the plot, theme, expression techniques, etc. in detail:
As you communicate more with AI, AI will get to know you more and more, know what you want, and even know you better than you do.
In the age of AI, it’s not about what you take on exams or what society asks you to do, but that you know what you love.
When choosing a professional field, prioritize love, not popularity.
Some professions that were once popular will most likely be replaced by AI in the future. Some majors that were unpopular in the past may shine in the future.
Whatever you do in the future, you can combine your personality with AI, and your experience, aesthetics, temperament, etc. are valuable assets.
The world is complex and changeable, and we must be curious about new things, constantly review and improve our learning methods, and build our own cognitive systems.
No matter how powerful AI is, the ultimate goal is still human.
The same AI tool, in different hands, will produce different results.
With the development of AI technology, the differences between tools are narrowing, but the differences between people are gradually widening.
The key to writing a good prompt is not to pursue complex techniques, but to clearly express the needs and context.
An insightful data analyst can use AI to create more compelling data visualizations, guide AI to generate more insightful data analysis reports, and transform AI outputs into more valuable business decisions.
Instead of passively using AI tools, people actively guide and unlock the potential of AI.
AI is strong when it is strong, and the stronger the human ability, the better the effect of AI application.
Many people use AI too simply, often initiating a task request once and not communicating deeply after getting the results.
Even when the AI gives an obviously wrong answer, it simply replies: Wrong, try again.
We can actually guide AI to reflect and stimulate its potential, for example: if your answer may be wrong, please analyze the part where it may go wrong, and then rethink and answer. Your answer ignores the market risk factor, please reconsider and add completely. Please check your reasoning process to identify possible logical loopholes. Please analyze what you just answered to find out where it may not be accurate. If you were asked to answer this question again, how would you improve?
Of course, the premise of guiding AI is that you can recognize the problem yourself.
AI is an amplifier of capabilities, just as a microscope amplifies humans’ ability to observe the microscopic world, telescopes expand their horizons to explore the universe, and AI amplifies humans’ ability to create and apply knowledge.
For some important and complex tasks, we can use plugins like ChatBrwal, which can be answered by multiple AIs at the same time by entering a question once, avoiding repeated copy-pasting and improving work efficiency.
For example, I have configured 4 commonly used AI tools: DeepSeek, Kimi, Doubao, and Yuanbao, just enter a question below, click send, and these 4 AIs will scramble to answer.
Next, I said to Gemini:
I’m completing an important and complex task: how to use AI at the highest level? I’ve asked 4 AI assistants to answer separately. Now that you are the judge, please formulate a set of 100-point grading rules according to the characteristics of the task, and then score the answers of these four assistants separately and explain in detail the reasons for your scoring.
Gemini replied as follows:
I then sent the 4 AI responses to Gemini, and it scored and commented on the rubric and explained the reasoning in detail.
I then said to Gemini:
Since you understand it so much, then answer this question yourself?
It will answer obediently, and I continue to ask:
Then you will give a score to your own answer according to the scoring rules you just mentioned, and explain the reason.
Gemini usually scores itself higher than other AIs, but generally not a perfect score, at which point you can continue to say to it:
So where are your remaining points? Please think about it and revise it again.
If you feel that Gemini’s ratings are not objective and fair, you can try switching to models like ChatGPT or Claude.
For us, what really matters is not the results of AI output, but the process of communicating with AI, which can gain some new ideas or inspiration.
AI is both an adversary that spurs us to progress and a friend who can help us amplify our strengths.
Humans choose to collaborate with AI as the best way to improve their cognition.
