The test was built to catch one small slice

When Alfred Binet designed the first practical intelligence test in the early 1900s, his purpose was narrow and honest: to find French schoolchildren who needed extra support. The tasks he chose rewarded the capacities that schooling rewards, working with words, reasoning through logic and number, holding information in memory and retrieving it on demand. The score that came out the other end, later packaged as IQ, was never meant to capture the full range of what a person could become. It measured a slice.

That slice turned out to be useful. IQ predicts fairly well how a person does in classrooms, and on tests that resemble classrooms, which is unsurprising given where it came from. The trouble began when the slice was mistaken for the whole.

The study that should have settled it

In 1921, the psychologist Lewis Terman began tracking more than a thousand California children with IQs above 140, following many of them for the rest of their lives (Terman & Oden, 1947). If a high IQ were the engine of remarkable achievement, this group should have produced a generation of history-makers.

They did not. Terman's subjects grew into healthy, capable, well-adjusted adults. They earned advanced degrees and built solid careers. Not one of them reshaped a field or left a lasting mark on the wider culture in the way a small number of their contemporaries did. Two California schoolchildren who were tested and turned away from the study, judged not bright enough to qualify, went on to become a Nobel laureate in physics and a President of the United States.

A second pattern points the same way. Average IQ scores rose steadily across the twentieth century, a finding so consistent it earned its own name, the Flynn effect (Flynn, 1999). Yet the rate of extraordinary contribution, the genuine breakthroughs in science, art, and leadership, did not climb alongside it. The number went up. The achievements it was supposed to predict stayed where they were.

Where the line breaks

Researchers who looked closely found that IQ stops being a useful guide surprisingly early. Above roughly 120, the correlation between intelligence scores and creative output essentially disappears (Getzels & Jackson, 1962). Past that threshold, additional IQ points buy very little in the domains where original work actually happens.

Leadership tells an even stranger story. The most effective leaders tend to have above-average intelligence, but not the highest. Beyond a certain point, a very high IQ can get in the way, making it harder to connect with the people one hopes to lead, and pulling a person toward abstract reasoning at the moments that call for reading a room.

Then there is the matter of profile. Sigmund Freud possessed enormous linguistic and psychological gifts alongside unremarkable mathematical ability and weak spatial skills. Picasso struggled to read, write, add, and subtract as a child. A single number averages these jagged profiles into a smooth figure that describes no one in particular, and predicts almost nothing about where a person will excel.

Extraordinary achievement is not a quantity sitting inside one person's head.

The deeper finding from decades of research is this. Extraordinary achievement emerges from the meeting of three things: a person, a domain they commit themselves to over years, and a field of people and institutions that decides what counts. Change any part of that system and the outcome changes. A test administered to a ten-year-old can see none of it (Moran & Gardner, 2006).

Why this matters now

For most of the last century, the limits of IQ were an academic concern. A person could overrate the test and still get along, because the capacities it measured, fast symbol manipulation, recall, converging quickly on the single correct answer, were genuinely valuable in school and at work.

Those are precisely the capacities that artificial intelligence now performs at scale and at speed. A machine recalls more, calculates faster, and converges on the textbook answer more reliably than any person sitting an exam. If the qualities IQ measures best are the qualities machines now handle, then ranking people by those qualities tells you less and less about what a person uniquely brings.

The human edge moves elsewhere. Toward asking the question no one thought to ask. Toward sensing what a colleague needs before they say it. Toward holding a purpose steady through uncertainty, and knowing oneself well enough to choose work that fits. These are real capacities, studied and trainable, and a single intelligence score was never designed to detect them.

IQ measured one kind of mind well. The age of AI is quietly revealing how much of human ability that one number left out, and how much of what it left out is the part that still belongs to people.

References & Further Reading

1. Conti, H. (2024). Multiple intelligences. EBSCO Research Starters. EBSCO Information Services.
2. Davis, K., Christodoulou, J., Seider, S., & Gardner, H. (2011). The theory of multiple intelligences. In R. J. Sternberg & S. B. Kaufman (Eds.), The Cambridge Handbook of Intelligence (pp. 485–503). Cambridge University Press.
3. Flynn, J. R. (1999). Searching for justice: The discovery of IQ gains over time. American Psychologist, 54(1), 5–20.
4. Gardner, H. (1983). Frames of Mind: The Theory of Multiple Intelligences. Basic Books.
5. Getzels, J. W., & Jackson, P. W. (1962). Creativity and Intelligence: Explorations with Gifted Students. Wiley.
6. Moran, S., & Gardner, H. (2006). Extraordinary achievements: A developmental and systems analysis. In D. Kuhn & R. S. Siegler (Eds.), Handbook of Child Psychology (6th ed., Vol. 2, pp. 905–949). Wiley.
7. Spearman, C. (1904). “General intelligence,” objectively determined and measured. American Journal of Psychology, 15(2), 201–293.
8. Terman, L. M., & Oden, M. H. (1947). The Gifted Child Grows Up (Genetic Studies of Genius, Vol. 4). Stanford University Press.