Originally written in Traditional Chinese: 記分板顯示不了 10 分：你給 AI 的框架，早就決定它能走多深
Montreal, 1976. Fourteen-year-old Nadia Comăneci scored the first perfect 10 in Olympic gymnastics. The scoreboard showed 1.00.
The judges were not wrong. The scoreboard had only three digits. Whoever designed it never imagined anyone would score a 10.
On day one of my Minerva Immersion Week in Tokyo, a speaker taught us to ask Japanese colleagues for a score from 1 to 5 with their fingers. Most show four, because with kaizen nothing is ever perfect. Japanese colleagues not showing five is culture deliberately leaving room. A scoreboard that cannot show 10 is imagination that left no room. Together they taught me this: the scale you design has already made half the decision for you.
The same is true when we talk to AI. The frame and constraints you give it decide, long before it answers, how deep it can go. Ask AI to grade a student's essay against a fixed rubric and it will be accurate. But if the student writes something brilliant that sits outside the rubric, can it see it? It can only find the nearest number on the scoreboard you gave it. That essay is its 10, and you handed it a three-digit board.
So let me be clear: AI is not in control. It can help you see many things you cannot. I use it every day. But its capabilities and its boundaries are, to a very large extent, decided by people: by the question you ask, the standard you set, how far you allow it to look. Too loose a frame and you get an average answer. Too tight and it cannot see the 10. However powerful AI becomes, it runs inside the range you draw.
This is not an argument against frames. Without a scoreboard, there is no competition. The point is to know the frame is yours, and that it may not fit what you most want to see.
Try this: before asking AI to score, rank or filter anything, ask yourself, "What is the highest number this board can show?" Then write down your own idea of a 10 first, and put it into the prompt: "Beyond this rubric, I also care about: ___." AI cannot see a scoreboard you never gave it. Only you can make the board bigger.
At Minerva, my master's is the MDA, Decision Analysis. The most important thing I learned there is to see clearly which scoreboard I am using.
References
- HISTORY.com, "Nadia Comăneci scores first perfect 10 in Olympic gymnastics" (July 18, 1976)
- Wikipedia, "Gymnastics at the 1976 Summer Olympics: Women's uneven bars"
Learn more about Minerva
- 👉 Minerva MDA (Master of Science in Decision Analysis): apply
- 👉 Minerva World Wise executive leadership program: express your interest
Follow in English
I also post in English on LinkedIn.