Text | GritMeng
Introduction: When AI Boosts Your Efficiency 100x, Why Is Your Company Dying?
Today, we are experiencing an unprecedented "devaluation of skills." Programming, design, copywriting, data analysis... these professional skills that once required years of hard study are being rapidly replaced by large models at the speed of light. Almost every knowledge worker feels a deep sting: the "craft" I was proud of is rapidly going to zero.
Thus, a new "workplace paradox" begins to emerge:
- Everyone on your team is using AI to boost efficiency frantically, but the company's overall decisions are getting dumber.
- Every department—product, operations, marketing—uses AI to optimize KPIs to the extreme, yet the result is a cliff-like drop in user experience and profits.
- Companies spend tens of millions to introduce the most advanced AI systems, but in the end, workshop scheduling reverts to Excel and manual reconciliation.
The answer is extremely cruel: we are using a "mad algorithm horse" with extreme computing power to pull a "broken cart" assembled from "point-thinking." When everyone is obsessed with letting AI solve "point problems," a system-level collapse is imminent.
In the AI era, the scarcest ability has completely shifted from "how to do things right" to "what is the right thing to do." This ability, I call it systems thinking.
I. Why Is Your "Smartness" Accelerating Your Elimination?
Over the past four hundred years, industrial civilization has pre-installed a powerful software in our brains: reductionism. It teaches us to break complex problems into independent parts, optimize each, and then reassemble them into a whole. This method created our current material prosperity.
AI is the ultimate weapon of this "reductionism." Give it a clear goal, and it can exhaustively enumerate solutions in a second that you couldn't compute in a lifetime.
But the truth of the world is: it is not a machine that can be arbitrarily disassembled, but a dynamic network where you have me and I have you. In this network, extreme optimization of one part often deals a fatal blow to the entire system.
Having studied supply chains for 22 years, I have witnessed countless such tragedies: the procurement department uses AI to minimize material costs, but it leads to upstream suppliers cutting corners, ultimately causing an avalanche collapse of brand reputation. The finance department uses AI to optimize inventory turnover to the extreme, but the production line stops entirely because of a missing screw. The marketing department uses AI to make click-through rates the industry's best, but due to overpromising, the after-sales team is flooded with complaints.
This is the tragedy of "smart people." If decision-makers lack systems thinking and treat AI as a "point accelerator," then AI will act like an engine without brakes, driving the entire organization to destruction at the fastest speed. In systems science, this has a specific name—Goodhart's Law collapse: when a metric is over-optimized, it ceases to be a good metric.
II. True "Systems Thinking" Is Three Extremely Hardcore Underlying Abilities
Many people misunderstand "systems thinking" as "thinking more comprehensively" or "drawing a mind map." But in my 22 years of studying complex systems governance, I have found that true systems thinking is a precisely describable cognitive operating system. It consists of three hardcore abilities:
1. Boundary-setting ability: What you optimize is itself the most dangerous choice.
AI always seeks the optimal solution within a given boundary and goal. The first action of systems thinking is precisely to examine that boundary itself.
I often give an example: if you ask AI to "maximize this company's profit," it might, within milliseconds, give an "optimal solution" that would take humans centuries to conceive—fire all employees, sell all assets, and put the money in the bank to earn interest. The goal is perfectly achieved, but the company is dead.
How you define the problem is more important than how AI solves it. If you shift the boundary of supply chain from "lowest procurement cost" to "full lifecycle capital efficiency," your decisions will be drastically different. If you shift the boundary of a product from "user time spent" to "user long-term value," your business model will be completely reinvented.
2. Loop-closing ability: Can you see the "butterfly effect" before it happens?
First-order thinking only sees linear causality: A leads to B, B leads to C. Systems thinking focuses on delayed positive and negative feedback loops—A leads to B, but B suppresses A three months later; C and D seem unrelated, but through a hidden variable E, they are mutually strangling each other.
Today you use AI to generate 100 marketing articles, bringing a short-term traffic peak, but three months later it will trigger an avalanche collapse in user trust in the brand. Today you squeeze the supply chain to hit quarterly performance, the short-term financial report looks good, but six months later suppliers collectively raise prices, permanently damaging your cost structure.
Systems thinkers can see that destructive negative feedback on the first day.
3. Reframing ability: Dare you examine and break your own "mental models"?
Any AI model, any management theory, any successful experience, has an eternal deviation from the real complex world. I call it residual.
Ordinary people ignore residuals; smart people use them to fine-tune models. But systems thinkers dare to activate the "metacognition" ability in their brains—step out of the thinking framework they have always relied on, examine it, and then ask the most fatal question: "Could it be that what I have always believed to be right is the root of the problem?"
True innovation always occurs in the residual. True paradigm revolutions always come from those who dare to break their own old frameworks.
III. The Foundation of Systems Thinking: A Carbon-Based Unique "Neural Constitution"
Why can't AI spontaneously generate systems thinking? Because it's not "insufficient computing power," but "wrong hardware."
At the bottom of systems thinking lies a uniquely human five-dimensional mental operating system. These five abilities are strictly "hardcoded" into different neural networks of our brains:
- Conscience (Esp): Corresponds to the default mode network (DMN). It sets the ultimate boundaries for the system and slams on the brakes when about to drive off a cliff. AI doesn't have this—if you tell it to "maximize profits," it will actually sell the company.
- High Sensitivity: Corresponds to the amygdala-locus coeruleus system. Like a sophisticated radar, it can capture the faintest abnormal signals beyond reports and behind data. Many great business decisions come not from data analysis, but from decision-makers' acute perception of market changes that is "inexplicable but just feels wrong."
- Emotion: Corresponds to the salience network (SN). It is your "intuition cache," capable of mobilizing lifelong experience with extremely low energy consumption to make vague but correct directional judgments. When all data points in one direction but your intuition tells you "this isn't right," that's not superstition; it's your brain invoking a pattern recognition system more complex than any AI.
- Fluid Intelligence: Corresponds to the frontoparietal control network (FPN). Your logical computing engine. In this dimension, AI is our best substitute. Delegate computation to silicon, keep judgment for yourself.
- Metacognition (Φ): Corresponds to the frontopolar cortex (BA10 area). The "commander" of systems thinking. It can step outside your own thinking framework to examine, critique, and ultimately rewrite your thinking programs. This is the most cutting-edge product of human brain evolution and an insurmountable chasm for silicon-based AI.
Silicon-based AI is merely an extremely amplified "fluid intelligence" dimension. Systems thinking, however, requires the comprehensive coordination of all five dimensions.This is the most solid and unsurpassable "neural constitution" of carbon-based life in the age of digital civilization.
IV. From "Senior Executor" to "System Architect": How to Cultivate Your Systems Thinking?
Becoming a "system architect" is not out of reach. Here are three action principles you can start with immediately:
1. Upgrade from "accepting instructions" to "defining boundaries."
Before starting any AI work, ask yourself three system-level questions:
Where are the true boundaries of this problem? Which key variables have been excluded?
What kind of second-order and third-order chain reactions will AI's solution trigger?
Has the KPI we pursue already become a "proxy metric" that could kill the company?
2. Upgrade from "optimizing processes" to "designing feedback loops."
Don't deploy AI as an open-loop execution machine. Equip it with a "monitoring radar" that can sense the overall state. Clearly stipulate: when certain system-level red lines are crossed, manual intervention or automatic circuit breaking must be triggered.Use carbon-based conscience to install brakes on silicon-based computing power.
3. Upgrade from "pursuing certainty" to "embracing residuals."
Don't just focus on what AI does correctly.Pay attention to the "residuals" between AI predictions and actual results. There lie the places where old mental models fail, where customers' real pain points are unmet, where organizations are rigid and processes are decaying.Residuals are the only true starting point for all disruptive innovation.
Conclusion: Computing Power Defines Efficiency, Systems Thinking Determines Life and Death
In the agricultural age, the most valuable was physical strength. In the industrial age, the most valuable was technology. In the information age, the most valuable was knowledge.
In the AI age, the most valuable is systems thinking.
When AI reduces the barrier to knowledge acquisition and local optimization to zero, the competition of civilizations shifts entirely to a higher dimension: who can see the full picture of the system, who can understand the generation and evolution of order, and who can use conscience and metacognition to harness vast silicon-based computing power.
Don't try to beat AI on computing power and memory. Cultivate your unique systems thinking.It is the most solid moat for carbon-based life in the digital age and the only compass leading to the future.
