I am not a superstitious person.
I believe in data. I believe in structured analysis. I believe in building arguments from evidence. If you have read any of my other posts, you already know that.
So when I tell you that I spent an afternoon having an AI analyze my Saju — the Korean fortune-telling system based on your birth date and time — and found myself genuinely unsettled by how accurate parts of it were, I want you to know I am saying that as someone who came in fully prepared to dismiss the whole thing.
I did not dismiss it.
First: What Is Saju?
Saju (사주) is a traditional Korean fortune-telling system with roots in Chinese astrology and philosophy. The name literally means “four pillars” — and those four pillars are built from the year, month, day, and hour of your birth.
Each pillar is represented by two characters: a Heavenly Stem and an Earthly Branch. Together, the eight characters (팔자, or “palja”) form a kind of cosmic fingerprint — a map of the energy patterns present at the exact moment you entered the world.
The system draws heavily on the Five Elements (Wood, Fire, Earth, Metal, Water) and the principles of yin and yang. A trained Saju reader — called a 역술가 — interprets the balance and interaction of these elements to describe a person’s innate character, the natural rhythms of their life, what kinds of energy they attract, and what periods of their life are likely to bring opportunity or challenge.
This is not the same as Western astrology, though people often compare the two. Western astrology focuses primarily on your sun sign — the position of the sun at your birth. Saju uses a far more specific calculation and places heavy emphasis on the hour of your birth, which means two people born on the same day can have very different charts depending on when they arrived.
In Korea, Saju is not fringe. It is deeply embedded in everyday life. People consult Saju readers before making major career decisions. Couples check compatibility before getting serious. Parents have their children’s charts read to understand what kind of support they might need. There are dedicated Saju apps in Korea with millions of downloads. The practice has been around for over a thousand years and shows no signs of fading.
The Traditional Way vs. The Modern Reality
Traditionally, you would visit a Saju reader in person. You would give them your birth information — ideally including the exact hour — and they would consult the charts, sometimes using physical reference books, and walk you through their interpretation over the course of an hour or more.
Good readers are in demand. Appointments are not always easy to get. The sessions are not cheap. And there is an inherent vulnerability in sitting across from someone and having them tell you things about yourself — you want to trust them, which means you are also somewhat at their mercy.
For people outside Korea, or for Koreans living abroad, access is even more limited. Your options are either to find a reader online (with all the uncertainty that comes with that) or to simply go without.
Which is where AI comes in.
Using AI to Read Your Saju
I tried this with Claude, and the experience was more substantive than I expected.
Here is what you need to provide:
- Your birth year
- Your birth month
- Your birth day
- Your birth hour (as specific as possible — even approximate is helpful)
- Your birth location (affects the time conversion in some interpretive traditions)
Once you provide that information, you can ask Claude to calculate your four pillars, identify your dominant and lacking elements, and interpret what that combination suggests about your personality, your strengths, your blind spots, and the general shape of your life’s energy.
What surprised me was the depth of the interpretation. Claude did not just spit out a generic horoscope. It explained the reasoning — why certain elements in my chart interact the way they do, what the historical significance of specific combinations means, and how those patterns might show up in real life.
I am not going to share the specifics of my reading here, because some of it felt genuinely personal. But I will say this: the part about how I approach work — the tendency toward structured thinking, the preference for frameworks over intuition, the drive to build systems rather than just react to situations — was described in a way that felt less like a horoscope and more like something a perceptive colleague might say after working with me for a few years.
That was not what I expected going in.
Is It Accurate? Here Is My Honest Take.
This is the question everyone asks, and I want to answer it carefully because I think the answer is more nuanced than a simple yes or no.
Some of it is going to feel accurate because it is describing broad human patterns — the tension between ambition and contentment, the ways certain personalities attract certain kinds of situations, the idea that different phases of life carry different kinds of energy. These things resonate because they are true for many people in some form.
But some of it is specific enough that the resonance feels less like a coincidence. The best Saju readings — even AI-generated ones — do not just describe who you are in flattering terms. They describe the shadow side too. The tendencies that create friction. The patterns you repeat. The kinds of challenges that tend to follow people with your particular elemental profile.
When a reading describes something uncomfortable about how you operate — and gets it right — that is harder to dismiss as a lucky guess.
My recommendation: go into it curious, not credulous. Do not make major life decisions based on an AI Saju reading. But do take the parts that resonate and sit with them. Sometimes the most useful thing a framework can do — whether it is a personality test, a therapy framework, or a thousand-year-old fortune system — is give you language for something you already felt but had not articulated.
How to Try It Yourself
If you want to try this, here is exactly how I did it:
Open Claude (or any capable AI) and start with something like this:
“I want you to analyze my Saju using traditional Korean four pillars fortune telling. My birth information is: [year], [month], [day], [hour]. Please calculate my four pillars, identify my dominant and lacking elements, and give me an interpretation of my personality, natural strengths, recurring challenges, and the general energy of different phases of my life. Please explain the reasoning as you go.”
The more specific you are with the hour, the better. If you do not know your exact birth time, give a range — “late morning,” “just after midnight” — and note the uncertainty.
From there, you can ask follow-up questions. Which years or age ranges does the reading suggest might be particularly significant? What element do I lack, and what does that tend to manifest as? What kinds of careers or environments tend to suit this elemental profile?
Treat it like a conversation, not a form. The more you engage with it, the more specific and useful the interpretation tends to become.
Why This Is Worth Your Time Even If You Are a Skeptic
Here is the thing I keep coming back to: Saju has survived for over a thousand years in a culture that is, by any measure, highly educated and highly practical. Korean society did not hold onto this system out of naivety. It held onto it because people kept finding value in it — in the self-reflection it prompted, in the language it gave them for talking about personality and life patterns, and in the way it encouraged people to think about their lives as having rhythms rather than just random events.
You do not have to believe the cosmos arranged your personality at birth to find that valuable. You just have to be willing to take the output seriously enough to ask yourself: does any of this ring true? And if so, what do I want to do with that?
I went in as a skeptic. I came out thinking it was one of the more interesting afternoons I have spent with an AI tool.
That is a lower bar than believing in fate. But it is a higher bar than I expected to clear.
Give it a try. The worst case is you spend twenty minutes and learn nothing. The more likely case is you end up with at least one thing worth thinking about.