In the face of the rapid development of AI, are we at the helm or are we the coerced passengers? In this era of “human-machine dance”, technology is no longer just a tool, but also quietly reshaping our way of thinking, decision-making logic and work rhythm. This article proposes 5 practical rules of human-machine coexistence to help you stay awake and take the initiative in the AI wave, and truly “control AI, not be controlled”.
Recently, chatting with friends, we all lamented that the era of AI is really coming! Before I knew it, AI had started to play a role in my various life paths:
- I went to the hospital to get the examination report, took photos and uploaded the bean bags as soon as possible, and helped me analyze the examination results;
- When encountering data processing problems in the work, the problem is described to deepseek, and analysis suggestions and processing formulas are given.
- An idea flashed in my mind, first talk to deepseek to sort out the ideas, and then give the cursor implementation page;
- If you want to read a book, send the book to Secret Tower AI and directly generate book PPT and explanation videos;
The great convenience brought by AI makes it easy to fall into these low-hanging pleasures.
However, under these pleasures, it is easy for us to develop a passive dependence on AI, and the inertia of thinking caused by this accessibility can make people fall into a deeper information cocoon.
In the AI era, we must not only learn to use AI, but also learn how to get along with it.
This article will explore the principles of collaboration with AI from five levels to build a positive, efficient, and autonomous human-machine collaboration relationship.
Principle 1: Maintain criticality and judgment
In the process of using AI, you must have encountered the situation of AI “nonsense”, and this “nonsense” is wrapped in a seemingly real logical chain, and this hidden illusion is not easy to detect.
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 >
For example, I asked AI to do product research on the ima knowledge base before, and the AI listed functional points such as “automatically generating to-do items”, but after actually experiencing the product, I found that there was no such function, and then I found that the information source was last year, because ima has iterated several versions, so there is a deviation in the information, and if this misleading information is directly adopted, it will output wrong judgments.
So the first principle of using AI is to have reasonable doubt and exercise your judgment:
- Question the source and context of the informationAfter the AI gives the answer, it can ask “what is the basis of this information” and “what are the limitations of this model”, etc., and let the AI self-check its answers through more in-depth questioning;
- Develop the habit of cross-validation: For key information and conclusions, they need to be confirmed by other reliable sources; When asking questions, you can also ask the AI to note its own information source;
- Strengthen your own knowledge base: The more professional knowledge in the corresponding field, the more accurate the identification of AI information accuracy; Or give the AI a designation
In short, we should approach the content output of AI with the mentality of “I am the person in charge of the final output, and AI is only the suggester”.
Principle 2: Actively collide with AI
A friend told me after using AI, “AI is nothing more than that”, I went to see his usage process, and in summary, it was “ask a vague question, accept a mediocre answer, and then end”, which is simply a great waste of AI’s potential.
I asked him to use AI as a dialogue partner for more collision of ideas, and he is now enjoying it.
For example, I discussed with AI about future interaction methods, let him play the role of an expert, and through asking questions from different angles, I had a clearer understanding of the problem and was able to put an idea floating in the air into a realistic bottle.
Talking to AI can start from these aspects:
- Iterative questioning: Constantly asking, in-depth, challenging, or asking for different angles in response to the AI’s initial response (“Why do you think so?”). “Is there a counterexample?” “What do you think [from a particular angle]?” ), through such questioning, to make yourself understand more deeply;
- role-playing: Let AI play a specific role to collide with opinions, “Suppose you are my entrepreneurial mentor, what advice would you make?” “Please criticize my views from the perspective of opponents”;
- Diversity and exploratory are required: “Please list three different ideas for achieving this goal” “Please analyze the potential pros and cons of this decision”.
I already like chatting with AI now, starting from an idea, there is something I don’t understand in the process, I can directly let AI explain, AI helps inspire ideas, and finally let AI summarize and output the chat content, perfect!
Principle 3: Take the initiative into your own hands
After getting used to using AI, there will be a strong path dependence, especially in complex tasks, over-reliance on AI may lose the ability to think independently, and it may also be limited by the “standard answers” provided by AI, and thinking will be framed.
I tried to collaborate with the AI in two ways to output content:
The first is that I only give one task and let the AI output the outline, and I enrich the content according to the outline;
The second is that I output the task and put forward some specific ideas, let the AI use these as the core to expand the outline, I modify it based on the outline, then let the AI enrich the content, I confirm the modification, and finally let the AI give some professional suggestions;
In practice, the second one will obviously be more in line with my requirements, because the kernel was invented, and I participated and confirmed each key step, and I had a stronger sense of control over the output content.
So, AI is a powerful tool, but it’s not our stand-in, don’t lose control of things and tasks.
To take the lead, you can start from these aspects:
- Set usage boundaries: The core must be produced by oneself, such as core ideas and ideas, which are the foundation of everything; Other scalable tasks such as information collection and sorting, first draft generation, divergence and aggregation, and summary can be handled by AI;
- Treat AI-generated content as a “draft” rather than a “finished product”: AI-generated content cannot be used immediately, it must be reviewed, proofread, and polished by you;
- Regular self-assessment: Regular review “After using AI, where do I save my time?” “Has my thinking ability become stronger or weaker?” “Do I still have control over myself?”
We must approach collaboration with AI with the mentality of “I am the leader, AI is the subordinate”.
Principle 4: Be a smart “lazy person”
The tools in the world are basically born for “lazy people”, and AI is the great artifact of our “lazy people”, but we must distinguish between positive and negative “lazy”:
- Positive “laziness”: Use AI to automate tedious tasks, such as data processing, formatting, basic data collection, grammar checking, etc., and set aside time for activities that are difficult for AI to replace, such as strategic thinking, creative ideation, in-depth reading, and interpersonal communication.
- Negative “laziness”: Outsource all thinking processes to AI, stop in-depth reading, no longer analyze information independently, and be satisfied with shallow explanations or “standard answers” provided by AI, which will eventually lead to a shrinkage of thinking ability.
For example, before I read a book, I will ask AI to help me generate the outline and introduction of the book, and then select the chapters I am interested in for in-depth reading, because “reading a book in 3 minutes” is never as good as thinking after in-depth reading.
AI has liberated the workforce, and we need to spend more time on higher-level thinking, creating, and decision-making.
Principle 5: Dance with AI
If you ask me “will AI replace humans”, I will answer “AI will be a part of humanity”, just like transportation expands the scope of human action and the Internet expands human information exchange channels, AI is also expanding human thinking ability, which is symbiosis rather than replacement.
To work with AI to solve problems and complete tasks, you can start from the following directions:
- Positioning complementarity: Clearly understand your own strengths and areas where AI is good at, and get used to thinking “What am I most valuable in this task?” What can AI help me share? ”
- Clarify the collaboration process: Disassemble reasonable workflow and division of labor according to the nature of the task;
- Positive feedback and debugging: Give specific feedback on the content given by AI, and in the long-term interaction with AI, you will find that AI often deviates from the main line, so you need to pull it back in time to guide AI to better serve your collaboration needs.
Let AI be part of our work, study, and life, because AI will do it even if you don’t want to.
The AI wave is surging forward, reshaping tools and the way we interact with the world.
These five principles remind us to never hand over the scepter of thinking, never extinguish the spark of curiosity, and never give up the search for meaning when enjoying the convenience brought by AI.
Let’s make good use of AI, but also our brains and hearts. Because in this era of intelligent gushing, the most scarce and precious thing will always be the product heart of “people” that remains sober, continues to create, and has the courage to take responsibility.
