In November 1960 the journal Science printed a paper with a startling title, "Doomsday: Friday, 13 November, A.D. 2026." Its authors were Heinz von Foerster, Patricia Mora and Lawrence Amiot, of the electrical engineering department at the University of Illinois in Urbana. They were not predicting a war or an asteroid. They were predicting arithmetic.
The three gathered 24 independent estimates of the world's population, covering about 100 generations from the year 0 to 1958. Ordinary exponential growth did not fit them, because humanity had been growing faster and faster. So they fitted a curve whose growth speeds up as numbers rise: population equals 179 billion divided by the years left until 2026.87, raised to the power 0.99. The fit was close, with an average error of about 7 percent, and the curve shoots to infinity at 2026.87, give or take 5.5 years. That point lands in mid November, and the authors chose Friday the 13th for it.
The paper is half joke and half warning. It notes that Charlemagne could have predicted the date within 300 years and Napoleon within 30. Its most quoted line says our great-great-grandchildren would not starve to death, they would be squeezed to death. Its proposed fix was a "peoplo-stat": cut the birth rate to about half, so the average family has just a little above two children.
For about three decades the formula did well. It gives 5.0 billion people for 1990, when the UN estimate is 5.3 billion. Then the gap opened fast. For 2020 the formula gives about 27 billion, and for 2024 about 63 billion. The UN World Population Prospects 2024 counts 8.2 billion in 2024.
The reason is births. According to the UN, women now have 2.25 children on average, down from 3.31 in 1990, and more than half of all countries are below the replacement level of 2.1. The UN expects world population to peak at around 10.3 billion in the mid 2080s and then slowly fall. The doomsday curve broke because the world ended up close to the family size the doomsday paper itself asked for.