It seems humanity is running out of children faster than expected. Fertility rates are collapsing around the world, often decades ahead of United Nations projections. Turkey’s fell to 1.48 last year—a level the UN thought would not arrive until 2100—while Bogotá’s is now below Tokyo’s. Even India, once assumed to prop up global demographics, has dipped under replacement. According to a new piece in The Economist, the world’s population, once projected to crest at 10.3 billion in 2084, may instead peak in the 2050s below nine billion before declining. (Among those experts mentioned, by the way, is Jesús Fernández-Villaverde, an economist at the University of Pennsylvania and visiting AEI scholar.)

From “Humanity will shrink, far sooner than you think” in the most recent issue: “At that point, the world’s population will start to shrink, something it has not done since the 14th century, when the Black Death wiped out perhaps a fifth of humanity.”

This demographic crunch has defied policymaker efforts. Child allowances, flexible work schemes, and subsidized daycare have barely budged birth rates. For its part, the UN continues to assume fertility will stabilize or rebound. But a demographer quoted by the magazine calls that “wishful thinking,” and the opinion is hardly an outlier. 

See if you find the UN assumption persuasive:

It is indeed possible to imagine that fertility might recover in some countries. It has done so before, rising in the early 2000s in the United States and much of northern Europe as women who had delayed having children got round to it. But it is far from clear that the world is destined to follow this example, and anyway, birth rates in most of the places that seemed fecund are declining again. They have fallen by a fifth in Nordic countries since 2010.

John Wilmoth of the United Nations Population Division explains one rationale for the idea that fertility rates will rebound: “an expectation of continuing social progress towards gender equality and women’s empowerment”. If the harm to women’s careers and finances that comes from having children were erased, fertility might rise. But the record of women’s empowerment thus far around the world is that it leads to lower fertility rates. It is not “an air-tight case”, concedes Mr Wilmoth.

Against this bleak backdrop, technology may be the only credible source of hope. Zoom boss Eric Yuan recently joined Bill Gates, Nvidia’s Jensen Huang, and JPMorgan’s Jamie Dimon in predicting shorter workweeks as advances in artificial intelligence boost worker productivity. The optimistic scenario goes like this: As digital assistants and code-writing bots shoulder more of the office load, employees reclaim hours for home life. Robot nannies and AI tutors lighten the costs and stresses of parenting, especially for dual-income households.

History hints at what could follow. Before the Industrial Revolution, wealth and fertility went hand-in-hand. That relationship flipped when economies modernized. Education became compulsory, child labor fell out of favor, and middle- and upper-class families invested heavily in fewer children’s education and well-being. 

But today, wealthier Americans are having more children, treating them as the ultimate luxury good. As AI-driven abundance spreads more broadly, perhaps resulting in the shorter workweeks those CEOs are talking about, larger families may once again be considered an attainable aspiration for regular folks rather than an elite indulgence. (Fingers crossed, given this recent analysis from JPM: “The vast sums being spent on AI suggest that investors believe these productivity gains will ultimately materialize, but we suspect many of them have not yet done so.”)

Indeed, even a modest “baby bonus” from technology would be profound. Governments are running out of levers to pull, dials to turn, and buttons to press. AI-powered productivity may not just be the best bet for growth, it could be the only realistic chance of nudging humanity away from demographic decline. This is something for governments to think hard about when deciding how to regulate this fast-evolving technology.