Ancient DNA on a massive scale shows human evolution has been speeding up, not slowing down
A landmark ancient-DNA study covering roughly 16,000 genomes has found that natural selection accelerated over the last 10,000 years, with the sharpest intensification concentrated around immune and metabolic traits. The finding challenges a common assumption that modern humans have largely escaped selective pressure.
The intuition that human evolution is essentially finished, that culture and medicine have replaced selection as the dominant forces shaping the species, has a long shelf life in popular science writing. David Reich, a population geneticist whose work draws on one of the largest ancient-DNA datasets ever assembled, describes a different picture. Natural selection on immune and metabolic traits, he argues, has not plateaued. It has picked up speed over the last 5,000 years relative to the 5,000 years that preceded them.
That claim now has substantial external support. The Broad Institute, reporting on a study built from roughly 16,000 ancient genomes, confirmed that natural selection has accelerated across the last 10,000 years of West Eurasian prehistory, with intensification specifically concentrated around immune function and metabolic regulation. The study draws on ancient-DNA samples spanning the period before and after the onset of farming, making it possible to compare selective pressure across two distinct eras rather than simply averaging across deep time.
The farming transition is the structural explanation that fits the data. As human populations shifted from mobile foraging to settled agriculture, they encountered new infectious disease burdens: denser communities, proximity to domesticated animals, stored food supplies. Metabolic demands also changed, with diets becoming more carbohydrate-heavy and less varied. Populations that adapted genetically to those conditions left more descendants. Reich’s framing, that selection “intensified” rather than merely continued, reflects what the genomic record shows: the rate of change in allele frequencies relevant to immune and metabolic function rose as the agricultural world expanded.
What we see is that this intensification of selection around immune traits, and similarly the intensification around metabolic traits, has accelerated over this time period. David Reich
What makes the finding non-obvious is precisely that the 5,000-year window in question sits within recorded and near-recorded history. This is not a story about archaic humans or the deep Pleistocene. The acceleration Reich describes covers roughly the period from early Bronze Age populations to the present. That is a timeframe modern people tend to think of as culturally rather than biologically eventful. The genomic evidence says both were happening simultaneously.
The scale of the underlying dataset matters to the credibility of the conclusion. Ancient-DNA studies with small sample sizes have historically been vulnerable to population-structure confounds: what looks like selection in one region may reflect migration from a population where a trait was already common. A dataset of 16,000 genomes, distributed across time and geography, provides the statistical resolution needed to distinguish genuine shifts in allele frequency from the noise introduced by population movement and admixture. Broad Institute reporting on the study describes this as a landmark precisely because earlier ancient-DNA work could not achieve that resolution.
The immune and metabolic findings are also worth treating as separate signals, even though they appear together. Immune trait evolution driven by pathogen pressure is conceptually familiar, though the speed and recency of the signal is striking. Metabolic trait evolution is in some ways more consequential for contemporary medicine. If selective pressure on metabolic function intensified as populations adapted to agricultural diets over the last several thousand years, then the mismatch between those evolved traits and modern dietary environments is not simply a story about recent industrial change. It has a longer evolutionary backstory.
Reich’s summary of the finding is direct: “What we see is that this intensification of selection around immune traits, and similarly the intensification around metabolic traits, has accelerated over this time period.” The external evidence from Broad Institute reporting and the underlying research literature does not soften or complicate that summary. It confirms it. The question the data opens is not whether recent human evolution is real but what it means that the most intense period of biological adaptation in the last 10,000 years coincides with the rise of the civilizations that produced writing, cities, and the idea that humanity had transcended its animal origins.