Start with one grain of rice on the first square of a chessboard and double it on every square, and the 64th square alone would hold more than 9.2 quintillion grains — with more than 18.4 quintillion grains across the whole board.

Sometime in the 13th century, a scholar named Ibn Khallikan wrote down a story that had probably been going around for centuries. A man invents the game of chess and shows it to his king. The king is delighted and offers any reward he likes. The inventor asks for something that sounds almost insultingly small: one grain of wheat on the first square of the board, two on the second, four on the third, doubling all the way to the sixty-fourth. The king agrees without a second thought. He should have done the math first.

In the older telling the reward is wheat, though people retell it with rice just as often. The names shift too. The inventor is usually called Sissa ben Dahir; the king goes by Shirham, Shahram, or Balhait depending on who is telling it.

What never changes is the ending. The king cannot pay.

Why the doubling gets away from you

The early squares are boring, and that is exactly the trap.

One, two, four, eight, sixteen. By the tenth square you are up to 512 grains, which fits in a cupped hand. Nothing about the first row hints at trouble. In one common retelling, the first bag of wheat only runs dry around the twentieth square. Twenty squares in, still just bags.

Then the ground opens up. Every square doubles the entire pile before it, so the amounts don’t add up steadily — they explode. Each new square is bigger than everything already counted.

The first half of the board, all 32 squares, comes to 4,294,967,295 grains. That is a lot of wheat, roughly 279 tonnes. But it is the whole first half, and it turns out to be almost nothing next to what comes later.

The two numbers that matter

This is where intuition gives out. The 64th square alone holds 9,223,372,036,854,775,808 grains. That single square carries more than two billion times what sat on the entire first half of the board. The whole board together comes to 18,446,744,073,709,551,615 grains, just over 18.4 quintillion.

Numbers this size stop meaning anything on their own, so it helps to weigh them. A full board of wheat grains would come to about 1,199,000,000,000 metric tons. For scale, the world is expected to grow around 842 million tonnes of wheat in 2025/2026. The chessboard total is over 1,400 times a full year of the global harvest.

Tell the story with rice, as many people do, and it gets no easier to pay. The world is projected to produce about 540.4 million metric tons of milled rice in 2025/2026. Whichever grain the king promised, no kingdom that has ever existed could have delivered it. The reward was never really a pile of food. It was a lesson dressed up as a favor.

Why our heads refuse to believe it

What keeps the puzzle alive is that people fall for it even when they know how it works. Martin Schonger and Daniela Sele, who studied how people estimate this kind of growth, found that people still underestimate it even when they are aware of the trap. Their more useful finding is that the mistake shrinks considerably when the growth is described as doubling times rather than percentage rates. The chessboard is essentially a doubling-time story told with a board instead of a clock, which is part of why it works.

The physicist Albert A. Bartlett put the human failing more bluntly, calling it “The greatest shortcoming of the human race is our inability to understand the exponential function.” Of course, that is a provocation, not a measured claim, and worth taking as one.

The underestimate is real and documented, though. A review of the same effect in disease spread notes that nonexperts underestimate exponential growth by assuming it moves in a straight line, a habit that mattered a great deal in the early spread of COVID-19.

The same trap shows up in newer places. Nathan Meikle, a business researcher at the University of Kansas, ran experiments on how people judge the pace of AI progress, and reached for the chessboard directly. “A simple example is would you rather have a billion dollars or would you rather have the money from doubling a penny 64 times?” he offered, the penny version running past $184 billion. From that work, Meikle argues that “We are, on average, going to be surprised at how quickly AI progresses and potentially surpasses human capability.” That is one researcher’s prediction, not a settled forecast. 

Also, no real doubling ever reaches the 64th square. Carl Sagan, writing about bacteria, put the limit plainly: “Exponentials can’t go on forever, because they will gobble up everything.” A real kingdom runs out of wheat, a real dish of bacteria runs out of room, a real technology runs into physics. The legend works precisely because it lets the doubling run unchecked to the end, which nothing in the world actually does.

That, we think, is why the story has outlasted the empires that told it. The grains were always beside the point. What the inventor understood, and what a king kept getting wrong across seven hundred years of retellings, is that people treat doubling as if it were simple addition. It isn’t, and the gap between those two ideas is wide enough to swallow a year of the world’s wheat many times over.