It’s not that AI is hard to use, but rather that your usage needs an upgrade.
“Why can others’ AI write code and create PPTs, while mine only says ‘I guess you might want to ask…’?”“The paper written with AI was criticized by my supervisor for lacking depth; is it just a high-level search engine?”
You may be experiencing these confusions as well. The truth is: while most people still treat AI as a “Q&A machine,” truly efficient learners have transformed it into a extension of thought and an amplifier of capabilities.
Today, we will not only discuss operational skills but also share a set of core mindsets that will truly enable AI to work for you.
Misconception: You May Have Fallen into the “AI Dependency Trap”
Many people’s usage of AI can be summarized into two extremes:
The first is “complete disbelief”: believing AI is just a toy, producing superficial content that cannot truly help them.
The second is “full acceptance”: throwing problems at AI and using all output results without any critical thinking.
Both approaches fall into the “AI dependency trap” — either rejecting tool advancement or abandoning independent thought. The truly smart individuals have found a third path:treating AI as a “thinking partner”.
Mindset One: Establish a “Dynamic Selection” Perspective — There is No Best Model, Only the Most Suitable Scenario
A common mistake in articles is to describe the capabilities of AI models as too static. In reality, this field is rapidly evolving on a monthly basis.
Practical selection guide for the current stage:
Doubao is like a quick-reacting intern: Its advantage lies in its familiarity with the domestic ecosystem, making it quick and convenient for handling daily tasks; its limitation is that it may lack depth in professional expertise. Suitable for data collection, drafting documents, and organizing meeting minutes and other daily scenarios.
DeepSeek is like a meticulous engineer: Its advantage is its outstanding coding and logical reasoning abilities, with high accuracy in responses; its limitation is that it may be weaker in creative expression and interdisciplinary knowledge. Suitable for programming debugging, mathematical calculations, and academic derivations that require precision.
ChatGPT is more like an all-around consultant: Its advantage is its strong comprehensive ability and clear thinking framework; its limitation is that its understanding of specific domestic scenarios may not be deep enough. Suitable for complex problem analysis, interdisciplinary thinking, and high-quality content creation.
Core point: Do not stick to one tool. True experts flexibly switch based on task types, even combining usage — using Doubao to collect local cases, using ChatGPT to build logical frameworks, and using DeepSeek to verify code snippets.
Mindset Two: Master “Conversational Creation” — There Are No One-Size-Fits-All Prompts, Only a Deepening Dialogue Process
The biggest misconception most people have when using AI is expecting to get a perfect answer with one question. This is like expecting to tell a chef “make a delicious meal” and getting a feast.
The real “prompt magic” is continuous dialogue:
First layer: Precise initiation — from vague “help me write a paper” to specific: “I will write a paper on ‘the impact of new media on teenagers’ reading habits,’ which needs to include empirical data, analysis of opposing viewpoints, and suggested measures. Please first provide me with three possible angles of discussion.”
Second layer: Iterative deepening — based on AI’s responses, engage in Socratic questioning: “Please expand on the second angle ‘the impact of short videos on deep reading ability,’ listing three sub-points and supporting evidence.”“What is the latest source of this data? Please provide authoritative references.”“What are the possible counterarguments to this viewpoint?”
Third layer: Critical examination — this is the key step that most guides overlook — actively seek out flaws in AI’s responses: “What methodological limitations might exist in the research conclusion you just mentioned?”“From a critical thinking perspective, what logical flaws exist in the argument you just provided?”
Core point: Treat AI as a thinking partner that needs your guidance, rather than a magic box that generates answers at the push of a button. Quality lies in the depth of dialogue.
Mindset Three: Always Maintain Judgment — AI is a Tool, You are the Subject
This is the most fundamental yet often overlooked principle in all AI usage.
Responsibility always lies with the user:
Fact-checking is essential: AI can “seriously mislead” (the phenomenon of hallucination), especially regarding data, citations, and facts in professional fields, which must be double-checked.
Boundaries of academic integrity must be clear: Understand the regulations regarding AI usage at your institution. There is a fundamental ethical difference between using AI to assist research and using AI to write papers.
Your expertise is the quality assurance: The deeper your understanding of the field, the better you can judge the quality of AI outputs and guide it to produce more valuable content.
Practice: Dual AI Workflow — Think Like a Professional
An efficient way to put the above mindsets into practice is to establish a dual AI workflow:
-
Planning phase (using Doubao/ChatGPT): Break down complex tasks and generate specific execution instructions: “I need to analyze the business model of shared bicycles; please generate a research outline that includes market size, competitive landscape, profit model, and user analysis.”
-
Execution phase (using DeepSeek/ChatGPT): Complete content creation in parts based on detailed instructions.
-
Verification phase (switching back to the first AI): Return the output for review: “Please critically review this business model analysis from an investor’s perspective, pointing out the three most likely weak points.”
This process simulates a professional working model: planning-execution-review, ensuring that the output has both breadth and depth.
Conclusion: Become the Master of Tools
AI is not here to replace you; it is here to extend your capabilities. It processes information, allowing you to focus on thinking; it provides materials, letting you take responsibility for creation; it offers options, enabling you to make decisions.
Starting today, try to interact with AI in new ways:
-
Next time you ask it to write copy, request it to provide three different styles at the same time;
-
When you ask it to analyze a problem, require it to provide arguments from both sides;
-
When you receive an answer, habitually ask, “What important factors might this analysis have overlooked?”
True experts are not those who know all the answers, but those who understand how to use tools to explore problems.
In this era where AI is ubiquitous, those who skillfully use tools will ultimately stand out.
Written by: Zhao Baichuan