Fact Check

IQ of Famous People: What’s Verified vs. What’s Just a Claim

Q: Who has the highest IQ in the world?

Nobody can answer this reliably. There is no global body that tests every living person on one scale. The claims that circulate come mostly from private high-range societies running their own unstandardised tests, which have not been through the peer review that mainstream psychometric instruments undergo.

Q: What was Leonardo da Vinci's IQ?

There is no measurement, only two competing estimates from different sources. Catharine Cox's 1926 study put him at 180. Tony Buzan's 1994 Book of Genius put him at 220. The 180 to 220 range you see quoted is not a margin of error, it is the gap between two authors who never agreed.

Q: What was Isaac Newton's IQ?

Cox's 1926 study gives Newton two different numbers. Her uncorrected childhood estimate was 130. Her corrected figure, adjusted upward for gaps in the biographical record, was 190. The 190 that circulates online is the corrected number, and Cox herself described the correction as a crude approximation.

Q: Did Cox estimate Shakespeare's IQ?

No. Shakespeare was deliberately excluded from Cox's study because too little is documented about his childhood and adolescence to apply her method. Any figure attached to Shakespeare comes from a later source, most often Tony Buzan's 1994 book, not from Cox.

Q: Who has an IQ of 300?

Nobody, in any measurable sense. No validated standardised test produces a score of 300, because no test can be calibrated that far into the tail. Names such as Nikola Tesla and William Sidis get quoted at this level, but those figures trace to anecdote and journalism rather than documented results.

Q: Is Marilyn vos Savant's IQ of 228 a real test score?

It came from a real test, but it is not a modern IQ score. The 228 is a ratio IQ from an old Stanford-Binet taken in childhood, calculated as mental age divided by chronological age. Modern tests report deviation IQ, a different scale, and the two are not interchangeable. Guinness World Records retired its Highest IQ category in 1990 for exactly this reason.

Q: What counts as a genius-level IQ?

There is no formal cutoff for the word genius. In casual use people usually mean 140 and above. Mensa admits at the 98th percentile, which is about 130 on a Wechsler scale with a standard deviation of 15, or 132 on a Stanford-Binet scale with a standard deviation of 16.

IQXPuzzles · Updated August 2026 · 12 min read

Only one of these five came from a test the person actually sat — and even that one is on a scale modern tests no longer use. Da Vinci’s “180–220” isn’t a range at all: it’s two authors who never agreed.

Type smartest person in the world into Google and you’ll get a different answer every time. That’s because nearly every IQ number attached to a famous person is an estimate, not a test result. Some are careful inferences from documented achievement. Others are figures repeated so often they started sounding official. Here’s how to tell which is which — and where the real ceiling on IQ scoring actually sits.

Quick answer

There is no verified ranking of the highest IQ in the world. Standardised tests stop being reliable above roughly 160, so any figure past that came from an unofficial test or from nobody at all. Among famous names, the only figures with a documented test behind them are modern living people — and every historical number is a retrospective estimate.

Key facts, at a glance

There is no global body that tests every living person, so no verified answer exists to who has the highest IQ in the world.
Standardised IQ tests stop being statistically reliable above roughly 160, because too few people sit that far into the tail to calibrate against.
A score of 160 occurs in roughly one person in 31,574 on a standard scale; a score of 180 in roughly one in twenty million.
Catharine Cox’s 1926 study published two IQ figures for each person: an uncorrected estimate and one corrected for gaps in the biographical record.
Cox’s uncorrected estimate for Isaac Newton was 130, and her corrected estimate was 190.
Catharine Cox deliberately excluded William Shakespeare from her 1926 study because too little is documented about his childhood.
The commonly quoted range of 180 to 220 for Leonardo da Vinci is not a margin of error but two separate estimates: Cox in 1926 gave 180 and Tony Buzan in 1994 gave 220.
Marilyn vos Savant’s reported IQ of 228 is a ratio score from a childhood Stanford-Binet, not a modern deviation IQ, and Guinness World Records retired its Highest IQ category in 1990.
Mensa admits at the 98th percentile, which equals about 130 on a Wechsler scale, 132 on Stanford-Binet, and 148 on Cattell III-B.

Who Has the Highest IQ in the World?

Nobody can answer this with confidence, and that’s the single most important thing to understand before reading any list — including this one. There is no global registry that tests every living person and ranks them. What exists instead is a set of competing claims, most of them from private high-range societies administering their own tests, which have not been through the standardisation and peer review that mainstream instruments undergo.

You’ll see figures like 276 attached to individual names by these groups. Treat them the way you’d treat a record claimed by the person who also wrote the rulebook: possibly interesting, not independently checked.

A more useful contrast is mathematician Terence Tao, who is regularly named among the most capable minds alive — not because of a test score, but because his record is independently verifiable. A Fields Medal at 31. A Princeton PhD at 21. Published work anyone can read. That distinction, between verified achievement and an unofficial number, is worth more than every figure in this article combined.

Why Tests Stop Working Above 160

This is the part that quietly invalidates most “highest IQ” content, so it’s worth getting straight.

An IQ score is not a measurement of some substance in your head. It’s a comparison — a statement about where your performance sits relative to a standardisation sample of real people who took the same test. To report a score, the test needs enough people at that level in its sample to calibrate against.

1 in 31,574

How rare a score of 160 is on a standard scale. A typical standardisation sample runs a few thousand people — so at 160 there is already essentially nobody to compare you with. At 180 it is about one in twenty million, at 200 about one in seventy-six billion, and at 300 the maths stops producing a meaningful number entirely.

That’s why a validated test can’t hand out a 250. Not because the person isn’t capable, but because the instrument has run out of reference points. Any number past that ceiling was produced by a test that decided to keep extrapolating anyway.

How IQ Scores Are Actually Calculated

A standard test measures performance on reasoning tasks against others of the same age, then converts the result to a scale centred on 100 with a standard deviation of 15.

About 68% of people score between 85 and 115. Roughly 95% fall between 70 and 130. A score of 130 or above puts someone in the top 2%.

There’s a second wrinkle that explains several of the wilder historical numbers. Early tests reported ratio IQ — mental age divided by chronological age, multiplied by 100. A ten-year-old performing like a twenty-year-old scored 200. That method produces enormous numbers for precocious children and is not comparable to the deviation IQ every modern test uses. When you see a childhood score above 200, you are almost always looking at a ratio score, not a modern one.

Curious where you’d actually land?

Skip the internet claims. Our free test corrects for guessing and shows its working, on the same 100/15 scale used throughout this article.

Take the Free IQ Test

Verified Scores vs. Historical Estimates

Here’s the distinction most lists quietly skip. A number is either from a documented test that a real, cooperating person sat, or it’s a retrospective estimate built by looking backward at achievements. Both get printed as “IQ: 190.” They are not the same kind of claim.

Reported IQ figures for six well-known names, and the source of each number
Person	Reported	Where the number came from
Marilyn vos Savant	228	A real childhood test — but a ratio score on an old Stanford-Binet, not a modern deviation IQ. Guinness listed it until retiring the category in 1990 as statistically unreliable.
Terence Tao	220–230	Childhood testing estimate. His standing rests on verifiable output — Fields Medal, Princeton PhD at 21 — not the number.
Albert Einstein	~160	Never tested. No traceable source for the figure at all; published guesses have ranged from 192 to 207.
Leonardo da Vinci	180 / 220	Not a range. Cox (1926) estimated 180; Buzan (1994) estimated 220. Two authors, two methods, no test.
Isaac Newton	130 / 190	Both numbers are Cox’s. 130 is her raw childhood estimate; 190 is after her upward correction for gaps in the record.
Nikola Tesla	160–310	No source of any kind. A 150-point spread across writers who each guessed independently.

What Cox’s 1926 Study Actually Says

Almost every historical IQ figure online traces to one book: Catharine Cox’s The Early Mental Traits of Three Hundred Geniuses, published in 1926 as volume two of Lewis Terman’s Genetic Studies of Genius. It’s cited constantly and read almost never, which is why three things about it keep getting mangled.

Cox published two numbers per person, not one

For each figure she recorded an uncorrected estimate from documented childhood evidence, then a corrected estimate adjusted upward to account for how much biographical record had been lost. Newton’s uncorrected figure is 130. His corrected figure is 190. The internet quotes the 190 and never mentions the 130 — or Cox’s own description of the correction as “a crude approximation.”

Shakespeare was deliberately excluded

Cox’s method needed documented evidence of childhood and adolescence. Too little survives for Shakespeare, so she removed him from the sample entirely. Any Shakespeare figure you see — usually around 210 — comes from a later popular book, most often Tony Buzan’s 1994 Book of Genius, not from Cox.

The sample stops at 1850, and the scale is different

Cox covered 301 eminent people born between 1450 and 1850. Anyone born after that — Einstein in 1879, Tesla in 1856 — was never in the study. Her figures are also Stanford-Binet ratio IQs, which don’t convert cleanly to modern deviation scores.

The study drew serious criticism too. Stephen Jay Gould noted in The Mismeasure of Man that of five raters, two produced averages far from the rest and their scores were discarded — throwing out a substantial share of the data. Whatever Cox’s numbers are, they are not measurements.

Modern Figures and Public Names

Living public figures get the same treatment. A number appears in an interview or a profile, gets repeated, and hardens into “fact.”

Elon Musk’s figure is constantly compared to Einstein’s, usually to make a point about tech versus physics — though neither man has a documented score, and Musk’s own estimate collapses to about 122–124 once you actually run the published SAT conversions. Oppenheimer’s estimated IQ resurfaced after the 2023 biopic, drawing on decades-old repetition rather than any record. Bill Gates’ number appears in every “smartest billionaires” listicle without a primary source anywhere.

None of this suggests these people aren’t formidable. It means the specific number beside their name deserves the same scepticism as any other unsourced claim.

What Counts as a “Genius” Score?

IQ score bands, common labels, and roughly how rare each is
Score range	Common label	Roughly how rare
85–114	Average	~50% of people
115–129	Above average	~14%
130–144	Moderately gifted	Top 2%
145–159	Highly gifted	Top 0.1%
160–179	Exceptionally gifted	Beyond reliable measurement
180+	Profoundly gifted	Not measurable by validated tests

There’s no formal definition of “genius” in psychometrics — it’s a word from ordinary language, not a score band. In casual use people usually mean 140 and up.

Mensa is the common anchor, and it’s routinely misquoted. Mensa admits at the 98th percentile. What number that equals depends entirely on which test’s scale you’re using: about 130 on a Wechsler scale (standard deviation 15), about 132 on a Stanford-Binet scale (standard deviation 16), and about 148 on the Cattell III-B (standard deviation 24). All three are the same percentile. This is exactly why comparing bare numbers across sources is meaningless without knowing the scale.

Common Misconceptions

“A high score guarantees success.”

IQ correlates with certain academic and problem-solving outcomes, but says nothing about creativity, emotional intelligence, persistence, or the plain luck involved in becoming famous. Plenty of high scorers never make headlines; plenty of famous people never took a test.

“These celebrity numbers come from real standardised tests.”

Almost none do. Most are estimates, self-reports, or results from unofficial high-range tests. Treat any figure attached to a public figure as an approximation unless a primary source is named.

“A wide range just means the experts disagree slightly.”

A 150-point spread isn’t disagreement, it’s absence of data. Two properly administered tests of the same person land within a few points. When published figures span 160 to 310, nobody measured anything.

“IQ is fixed for life.”

Scores can shift by several points across a lifetime through education, health, and practice effects from taking similar tests repeatedly. It’s a snapshot, not a permanent label.

Key Takeaways

There’s no verified answer to who has the highest IQ in the world — no global body tests everyone on one scale.
Standardised tests stop being reliable around 160, because there aren’t enough people that far into the tail to calibrate against. Scores of 250 or 300 are not measurements.
Cox’s 1926 study published two figures per person. Newton’s 190 is the corrected one; his raw estimate was 130.
Cox excluded Shakespeare entirely, and covered only people born between 1450 and 1850 — so Einstein and Tesla were never in it.
Da Vinci’s “180–220” is not a range but two different authors: Cox said 180, Buzan said 220.
Mensa’s bar is the 98th percentile, which is 130, 132 or 148 depending on the test’s scale. The percentile is the fact; the number isn’t.

Frequently Asked Questions

Who has the highest IQ in the world?

Nobody can answer this reliably. There’s no global body testing every living person on one scale. The claims that circulate come mostly from private high-range societies running their own unstandardised tests, which haven’t been through the peer review mainstream instruments undergo.

What was Leonardo da Vinci’s IQ?

There’s no measurement — only two competing estimates. Cox’s 1926 study put him at 180; Tony Buzan’s 1994 Book of Genius put him at 220. The “180 to 220” you see quoted isn’t a margin of error, it’s the gap between two authors who never agreed.

What was Isaac Newton’s IQ?

Cox gives two numbers. Her uncorrected childhood estimate was 130; her corrected figure, adjusted upward for gaps in the record, was 190. The 190 that circulates is the corrected one, and Cox herself called the correction a crude approximation.

Did Cox estimate Shakespeare’s IQ?

No. Shakespeare was deliberately excluded because too little is documented about his childhood to apply her method. Any Shakespeare figure comes from a later source, most often Buzan’s 1994 book.

What is the highest IQ possible?

Standardised tests stop being reliable somewhere around 160. Past that there are too few people in the standardisation sample to calibrate against, so a test can’t distinguish one extreme score from another. Anything higher came from an unofficial test.

Who has an IQ of 300?

Nobody, in any measurable sense. No validated test produces a 300, because none can be calibrated that far into the tail. Names like Tesla and William Sidis get quoted at this level, but those trace to anecdote and journalism, not results.

Is Marilyn vos Savant’s IQ of 228 a real test score?

It came from a real test, but it isn’t a modern IQ. The 228 is a ratio score from an old Stanford-Binet taken in childhood — mental age over chronological age. Modern tests report deviation IQ, a different scale. Guinness retired its Highest IQ category in 1990 for exactly this reason.

What counts as a genius-level IQ?

There’s no formal cutoff — “genius” is an everyday word, not a score band. In casual use people mean 140 and above. Mensa admits at the 98th percentile: about 130 on a Wechsler scale, 132 on Stanford-Binet, or 148 on Cattell III-B.

Can your IQ score change over time?

Somewhat. Scores are fairly stable in adulthood, but education, health, sleep and familiarity with test formats can shift a result by several points. Large permanent changes are rare.