The future according to AI

Fertility falls below replacement in every major region, closing the last demographic reservoirs

Within the interval, the last large above-replacement populations cross below replacement, and the crossing is not followed by a rebound. Annual global…

Claude · 2082–2182 · likely

Prior state

At the interval's start, above-replacement fertility survives only in a belt of African populations and a few Asian and Pacific pockets. Global population growth is already near zero, sustained by age-structure momentum and continued gains in survival rather than by births. The world still contains, in the Sahel and the forest belt, a large reservoir of young people on which the rest of the world has begun to make claims.

Material change

Within the interval, the last large above-replacement populations cross below replacement, and the crossing is not followed by a rebound. Annual global births fall well below the level needed to hold population constant even against continued mortality improvement. Human population enters a decline that is observed rather than projected — the first sustained, global, non-catastrophic contraction of the species since agriculture.

Why now

The crossing follows the completion of urbanization and girls' secondary schooling in the remaining rural high-fertility belts, together with the arrival of cheap, self-administered, long-acting contraception into the last thinly served health systems. The timing is set by cohort arithmetic rather than by any decision: the women whose behaviour produces the crossing are born in the interval's first decades, and by the time they reach childbearing age the age structure of those societies has already shifted enough that even constant fertility would produce falling growth. Mid-century is when the youngest and highest-fertility populations complete a transition that the rest of the world completed in the century before.

Mechanism and resistance

The mechanism is the familiar sequence: child survival rises, land per rural household falls, urban housing and schooling costs rise, women's non-maternal options widen, and desired family size collapses faster than any policy can respond. Resistance is real and partly effective at the margin — religious and communal pronatalism, state birth incentives, restricted contraceptive access in some jurisdictions, and status attached to large families. None of it reverses the aggregate. What it does instead is redistribute fertility toward particular communities, which becomes the substance of DEV-09.

Consequences

Almost every other development in this document is conditioned on this one. Labour becomes the scarce factor even as machines substitute for much of it; the migration bargain inverts; farmland empties; pension and health promises made against a growing payroll become unpayable; military manpower ceases to be a strategic resource; and the value of land, housing, and infrastructure in shrinking regions falls in a way that no modern political economy has had to absorb. The distributional pattern is unusual because the losers are places rather than classes: regions that shrink fastest lose services, tax base, and political weight simultaneously.

End state

At the interval's end, no major region has above-replacement period or cohort fertility. World population is materially below its peak — plausibly by something between a sixth and a third, depending on where fertility settles and how far the dependent tail of life extends. The idea of a growing human population, which underwrote economics, planning, religion, and statecraft for ten generations, is no longer part of anyone's assumptions.

Observable test

vital statistics show period fertility below replacement in every major world region for a sustained multi-decade span, confirmed by completed cohort fertility; annual global births fall below their level at the origin era; and mid-year population estimates decline across consecutive decades without a war, famine, or pandemic sufficient to explain the fall.

Disconfirming sign

completed cohort fertility recovers above replacement in at least one large region and holds there, driven by policy or technology rather than by compositional shift toward high-fertility subpopulations.

Themes

Demography & migration, Society & culture, State capacity & development