I haven´t double-checked this, but it sounds about right. I´ve read something similar in books a couple of years back. This a propos that confused young man from South Korea who claims to be "the world´s IQ record holder" at IQ 276. As I expected, such a high IQ score is simply impossible. And that´s that.
Chris Langan also claims to have a very high IQ (about 200), but he has at least constructed a complex metaphysical system. YoungHoon Kim simply repeats all the usual amateur apologist arguments for Christianity we have heard thousands of time before. In fact, he does so with such great conviction - as if he was saying something new or smart - that he becomes almost a parody of the entire field. His Dunning-Kruger is certainly off the charts...
And yes, I know this is worthless knowledge, but I just can´t help myself...
Short answer: claims like “IQ = 276” are not just unlikely — they are mathematically meaningless.
Let’s unpack why.
1. What IQ scores actually are
Modern IQ tests (like the Wechsler Adult Intelligence Scale or the Stanford-Binet Intelligence Scales) do not measure intelligence on an absolute scale.
They are norm-referenced:
where is how many standard deviations above or below the population mean a person is.
So:
-
IQ 100 = average
-
IQ 130 = 2 SD above mean (≈ top 2%)
-
IQ 145 = 3 SD
-
IQ 160 = 4 SD
…and so on.
But here’s the crucial point:
IQ tests are only calibrated and validated for a limited range of -scores.
Beyond that range, the numbers lose meaning.
2. Why extreme IQ numbers collapse
To assign a real IQ of, say, 200, a test would need to:
-
Contain items hard enough to distinguish people at ~7 SD above the mean
-
Have a large enough norming sample to estimate how rare such performance is
-
Demonstrate reliability at that level
None of this exists.
For perspective:
-
5 SD (IQ ≈ 175) → 1 in 3.5 million
-
6 SD (IQ ≈ 190) → 1 in 500 million
-
7 SD (IQ ≈ 205) → 1 in ~80 billion (more than the number of humans who have ever lived)
So if someone claims an IQ of 276:
That corresponds to a rarity of about:
1 in 10³¹ humans
That is more than the number of atoms in a cup of water.
It’s not “very rare.”
It is statistically impossible in any meaningful sense.
3. What IQ tests actually top out at
Most well-designed modern IQ tests:
-
Can measure reliably up to about IQ 160–170 (≈ 4 SD)
-
Can give rough estimates up to about IQ 180 (≈ 5 SD)
Beyond that:
-
Ceiling effects dominate
-
Scores depend on extrapolation, not measurement
-
Small errors in raw score become enormous errors in IQ
This is why official manuals usually cap reporting around 160 or 165.
Anything above that is basically:
“You hit the ceiling; you are smarter than this test can tell.”
4. Where do numbers like 200–300 come from?
They usually come from:
-
Untimed or experimental tests never properly normed
-
Extrapolating from childhood scores
-
Using ratios instead of deviation IQs
-
Self-administered online tests
-
High-IQ societies inventing their own scales
These violate basic psychometrics.
The numbers look impressive — but they are numerology, not measurement.
5. The highest IQ that can be meaningfully claimed
In real psychometrics:
| IQ | Meaningfulness |
|---|---|
| ≤145 | Very solid |
| 145–160 | Strongly reliable |
| 160–170 | Still meaningful |
| 170–180 | Rough, ceiling-limited |
| >180 | Not psychometrically real |
So the highest meaningful IQ is about:
~170–180 (4–5 SD above the mean)
Everything above that is not “more genius” — it is outside the measuring instrument’s domain.
6. Why this doesn’t diminish genius
People like John von Neumann or Kurt Gödel may well have had cognitive abilities far beyond most test ceilings — but IQ cannot quantify that.
At the extreme high end, intelligence becomes multidimensional, and a single number breaks down.
Bottom line
When you see:
“IQ = 240” or “IQ = 276”
what it really means is:
“Someone took a test that could not measure them, and a big number was invented.”
The real upper bound of meaningful IQ is around 170–180 — after that, the scale stops working.
