Opens in a new window Opens an external website Opens an external website in a new window
This website utilizes technologies such as cookies to enable essential site functionality, as well as for analytics, personalization, and targeted advertising. To learn more, view the following link: Privacy Policy
Main content starts here
ScienceInsider
Math
8 Oct 2026
2:00 PM ET
An apartment building in Hong Kong. In 2026, researchers predicted 66 years ago, there would be an infinite number of people on Earth.Marcio Rodrigo Machado/S3studio/Getty Images
1.0x
00:00
6:28
1.0x
“A physicist has put a definite date on Doomsday,” The New York Times wrote on 4 November 1960. “He calculates that it will come on Friday the Thirteenth in November, 2026 A. D. On that date, or thereabouts, the human population will have expanded so far as to approach infinity and thus will automatically annihilate itself.”
The newspaper was right—but the prediction was wrong. That week, Austrian American physicist Heinz von Foerster and two colleagues published a paper in Science with the stunningly succinct title “Doomsday: Friday, 13 November, A.D. 2026.”
The work was based on a simple observation: The time it took for the human population to double in size kept getting shorter. Using population estimates from about 400 B.C.E. until 1958 C.E., the authors—all from the department of electrical engineering at the University of Illinois Urbana-Champaign—came up with an equation that best described the trend. If things continued the way they were going, the number of humans on Earth would rise faster and faster, reaching more than 25 billion by 2020.
The researchers didn’t calculate when humanity would run out of food, land, or energy. They just concluded that at some point this year, the equation ceased to give meaningful results because the predicted population shot toward infinity. That was their doomsday. “Our great-great-grandchildren will not starve,” they wrote. “They will be squeezed to death.”
The exact date was somewhat arbitrary, however. The formula solved to 2026.87 plus or minus 5.5 years, and The New York Times reported that von Foerster “searched the center of this doom-time span for an appropriate date and found that a Friday the Thirteenth would fall conveniently in November of 2026.” (13 November also happened to be his birthday.)
The paper made a lasting impression on Joel Cohen. A mathematical population biologist at Rockefeller University, he says he remembers hearing about it as a high school student. “I didn’t understand the mathematics and I didn’t understand the demography, but it made a splash.”
Social scientist Helga Nowotny, who works at the Complexity Science Hub, an independent research center in Vienna, remembers reading the study sometime after it came out and finding it intriguing. “I said: ‘OK, I’ll keep this and hopefully I will live long enough to see whether it happens or not.’”
But demographers, annoyed by what they saw as a provocation by outsiders, pointed out obvious flaws. For example, the time it takes for the number of humans to double could get a lot shorter, but certainly not shorter than the 9 months it takes for a baby to develop. “In view of the comments that subsequently were published in this journal, an extensive discussion of this article does not seem to be required,” public health expert Harold Dorn drily noted in Science 2 years after von Foerster’s article was published. The forecast would set a record, he predicted, “for the short length of time required to demonstrate its unreliability.”
Indeed, von Foerster’s approach didn’t make much sense, says Patrick Gerland, who heads the Population Estimates and Projection Section at the United Nations. Rather than fitting some overall curve to global populations, demographers generally model individual country populations using data on births, deaths, and migration. Then they add up the totals. “There is a lot of diversity in the world,” Gerland says. “Many countries don’t experience the same situation in terms of socioeconomic characteristics and development.”
And yet the formula held up reasonably well—for a while. By 1975, the world population had reached about 4 billion, even more than the roughly 3.6 billion predicted by von Foerster’s formula.
But in the early 1990s, after fertility in rich countries and China began to drop, the formula proved too pessimistic. By 2000, the world population was closer to 6 billion than the 7 billion von Foerster and colleagues had predicted; by 2010 it was at 7 billion rather than 11 billion. Now, 1 month before von Foerster’s doomsday, there are about 8 billion people—not hundreds of billions.
But that’s beside the point, Cohen says. “I don’t think [von Foerster] was seriously trying to predict the future,” he says. “He was drawing attention to a pattern which I don’t think was sufficiently appreciated.” Indeed, the researchers themselves wrote they hoped to “add some fuel to the heated controversy about whether or not the time has come when something has to be done about population growth control.”
The irony is that von Foerster was pointing out a pattern of faster than exponential growth just as it was coming to an end, Cohen says. In fact, the world population only grew in a way resembling his formula for a short period, economist Javier Birchenall of the University of California, Santa Barbara writes in a Perspective published in Science today. “The failure of the Doomsday equation is ultimately a cautionary tale about the dangers of extrapolating from historically exceptional circumstances,” he writes.
The paper has kept intriguing people. Victor Yakovenko, a physicist at the University of Maryland, recently revisited it and modified the formulas so that instead of continuing into infinity, the human population reaches a peak and then starts to decline. He predicts this peak will arrive around 2030, which Gerland calls “implausible.” (Most demographic models don’t predict a decline before 2080.)
Nowotny—happy she lived long enough to see how the story ends—is co-organizing a conference in Vienna on 13 November “on the sense and nonsense of end-of-world predictions.” The dire artificial intelligence (AI) doomsday scenarios currently circulating have made the topic more timely, she says.
Indeed, the point at which the doomsday equation blows up is called a “singularity” in math and von Foerster used that term. Today, another “singularity” is discussed widely in Silicon Valley, as the hypothetical moment when a self-improving AI leads to runaway acceleration of technological progress that radically transforms society—or ends it.
Forecasting this singularity in AI, too, depends on extrapolating current trends into the future, and fitting a curve to often cherry-picked data. And like von Foerster, those who make such predictions may turn out to be completely wrong. Or maybe not.
doi: 10.1126/science.zzhv2dd
MathScientific Community
U.S. crop genome sequencing gets $180 million boost
By
Erik Stokstad
Department of Energy expands ‘Genesis Mission’ to boost AI, awarding larger grants
By
Adrian Cho
Major employer of archaeologists shuts down, leaving workers in the lurch
Amid the unprecedented upheaval in U.S. science, hard-hitting, trustworthy reporting has never been more important. As part of the AAAS mission, Science has built a global award-winning network of reporters and editors that independently cover the most important developments in research and policy.
Your tax-deductible contribution plays a critical role in sustaining this effort.
If, at any time, you would like to stop or make changes to your monthly gift, please contact the Philanthropy Office at philanthropy@aaas.org or 202-326-6636.
Help News from Science publish trustworthy, high-impact stories about research and the people who shape it. Please make a tax‑deductible gift today.