James here, CEO of Mercury Technology Solutions. Taipei, Taiwan — May 29, 2026
I recently received a message from my friend asking about how to cultivate independent judgment in children. She listed various parenting techniques and asked if these were the right ways to build a child’s character, worrying that giving them too much autonomy might lead to disaster.
First of all, stop worrying about these microscopic details.
Let me ask you a much bigger question: If you could time-travel back to 1980, would you make sure your child took the university entrance exam? If you went back to 2000, would you force your child to buy real estate in a Tier-1 city?
You wouldn't worry about whether studying in 1980 distracted them from feeding the family sheep. The sheep will figure out how to eat; it won't die. But if your child missed that macroeconomic window, you would regret it for the rest of your life.
This is the most critical concept you must understand right now: You must identify the macroeconomic context of your era.
1. The Devaluation of "Experience"
For the past several decades, what exactly were you selling to the labor market? What made a human being valuable?
Most people say: "Experience."
But "experience" is a vague term. We need to split it into two distinct functions:
- Empirical Data Analysis
- Empirical Outcome Judgment
What exactly is Empirical Data Analysis? It is statistical work.
A few years ago, when AlphaGo defeated Lee Sedol, I told my readers that the world had fundamentally changed. How does a human play Go? Humans play a game, record the moves (the data), study the strategies of past masters, and try to innovate within those boundaries. A human operates by analyzing a limited dataset.
The number of possible board configurations in Go is astronomical. A human brain cannot compute all of them. How did AlphaGo win? Brute force exhaustion. AlphaGo does not "understand" Go. It simply uses massive computational power to statistically analyze a dataset far larger than any human could ever comprehend.
Generative AI (like OpenAI’s models) operates on the exact same principle. It does not "understand" truth. It simply uses statistics to predict the most probable next token across the entire internet.
The harsh reality is this: In the realm of Empirical Data Analysis, human value has plummeted to zero. Humanity invented compute power just like we invented electricity. Once the electric train was invented, humans stopped pulling carts. Now that we have cheap AI tokens, humans no longer need to perform limited-scope data analysis.
Take driving, for example. What makes an experienced driver? They have accumulated 10 years of road-condition data in their brain. But how is autonomous driving trained? Millions of drivers upload their daily data to the cloud. The AI synthesizes this and instantly gains millions of years of driving experience across every road on Earth. It becomes the AlphaGo of driving.
If you try to compete with AI in Empirical Data Analysis, you will lose. You can grind yourself to the bone trying to be the "Lee Sedol" of your specific industry, and you will still be utterly worthless to the market.
2. The Premium on "Outcome Judgment"
If Empirical Data Analysis is dead, your market value in 2026 relies entirely on the second function: Empirical Outcome Judgment.
This is your "Buying Real Estate in 2000" moment.
Now, we face two critical questions. Question 1: Can a person make good Outcome Judgments if they have zero foundational experience in Data Analysis?
Absolutely not. As the ancient philosopher Han Fei said: "Fierce generals must rise from the infantry; prime ministers must rise from the local provinces." Emperor Liu Bang (founder of the Han Dynasty) was not as good at administration as Xiao He, not as good at strategy as Zhang Liang, and not as good at warfare as Han Xin. But his skills in those three areas were not zero. If he had zero foundational knowledge, he wouldn't have been able to identify those three geniuses, command them, or trust their judgment.
Therefore, while Empirical Data Analysis is no longer your primary job, it must still be your foundational baseline.
Question 2: Does having foundational Data Analysis skills guarantee you will make good Outcome Judgments?
Sadly, no. This is what we call "having independent judgment." It is a character trait, not just a skill. Experienced people often lack judgment. So, how do you cultivate it? How do you become decisive?
You become decisive by executing thousands of decisions.
3. The Simulation Strategy (Time and Cost Compression)
This is where we solve the reader's dilemma about cultivating judgment.
To make thousands of decisions in the real world requires two things most people don't have: Time and Money. If you make a bad business decision in the real world, who pays for the bankruptcy? If a decision takes ten years to yield a result, how can you learn fast enough?
The solution is simple: Compress Time and Lower the Cost.
If the owner of a traditional car factory wants his son to learn how to build the business from scratch, he can't. The historical data from 30 years ago is gone.
But as someone in high-frequency trading, can I teach my son how I built my career? Easily. I can open a digital simulation account for him, give him $30,000 in fake capital, load up the real historical market data from 2008, and set the simulation speed so that one full trading day passes in one minute.
In reality, a specific arbitrage opportunity might only happen once a year. But in the simulation, he encounters a small error every 5 minutes and a massive one every 3 hours.
He will make mistakes. He will misidentify an error, go all-in, and blow up his account. It costs him nothing but an hour of time. We review the data. Why did he blow up? Because he didn't hedge his position across a secondary exchange. He runs the simulation again. This time, the exchange's servers crash, trapping his capital. He blows up again. He runs it again, and again, and again.
After 50 hours of intensive simulation, he hits the ceiling of the market's liquidity. He turns his fake $30,000 into millions.
Did he make any real money? No. But is the boy who finished those 50 hours the same boy who started? Absolutely not.
He has developed brutal, decisive judgment. He knows exactly how the market reacts because the simulation was built on real, historical data. Those 50 hours of compressed simulation gave him the equivalent of 10 years of real-world trading scars.
If you spend your time running these compressed simulations—whether in trading, business models, or AI workflows—you will develop a massive comparative advantage in Empirical Outcome Judgment.
When your competitor walks into a room, they boast about having 10 years of experience. When you walk in, you possess 5,000 years of experience—4,990 years of simulated data processing, and 10 years of real-world application.
Pair that decisive judgment with an army of AI Agents working 24/7, and the value you bring to the market will exceed that of 100 human employees. If you charge the salary of 30 employees, what company could possibly refuse to hire you?
That is how you increase your market price. And when your price goes up, everything else in the world suddenly becomes cheap.
Mercury Technology Solutions: Accelerate Digitality.
