Full Breakdown
AI Reveals Explosive Evolutionary Bursts in Birds Over 45-Million-Year History
7/22/2026, 11:37:20 AM
Core Findings
An artificial-intelligence analysis of passerine (songbird) skeletons shows a pronounced burst of body-shape evolution roughly 35 million years ago, coinciding with the Eocene-Oligocene Transition—a rapid shift from a warm “greenhouse” climate to a cooler “icehouse” world. The same analysis identifies a subsequent slowdown about 15 million years ago, when most ecological niches had already been occupied. These patterns support the long-standing idea that new groups often arise in brief, rapid radiations followed by extended periods of relative stasis.
Methodology and Data
Researchers measured a dozen skeletal bones across more than 2,000 passerine species, generating over 170,000 quantitative observations—the largest dataset of its kind for this order. The measurements were fed into a newly developed AI model capable of detecting subtle shape changes across deep time. By tracking these changes throughout the group’s 45-million-year history, the model pinpointed the timing and magnitude of evolutionary rate shifts.
Evolutionary Pattern Context
The observed “burst-then-slowdown” sequence aligns with a century-old hypothesis that evolutionary radiations are triggered by new ecological opportunities—such as the opening of habitats during major climate transitions—or by dispersal to new continents. As niches fill, the pace of morphological innovation declines, producing the long quiet intervals seen after the 15-million-year slowdown.
Expert Commentary
University of Michigan researchers emphasized the significance of the findings for evolutionary theory. Jake Berv highlighted that the pattern mirrors predictions made a hundred years ago about explosive diversification during radiations. Brian Weeks noted that the alternating spikes and declines in evolutionary rates reflect lineages exploiting newly available ecological space. Both scholars argue that AI’s ability to sift massive datasets uncovers patterns invisible to the human eye, underscoring a promising direction for future paleobiological research.
Publication and Implications
The study appears in *Nature Ecology & Evolution*. Its authors suggest that AI-driven analyses could reshape how scientists investigate macroevolutionary dynamics, offering a powerful complement to traditional fossil-record interpretations.
Verbatim Quotes
- “This is really important for evolutionary theory because there's a long history, going back 100 years, that predicts the emergence of new groups, called evolutionary radiations, is often associated with an explosive burst of diversification,” — Jake Berv, at the University of Michigan, in a statement
- “Our findings have definitely shifted my thinking about how the world works,” — Brian Weeks, also at the University of Michigan
- “This pattern we found with rare, big increases in the rates of evolution and lots of small decreases in the rate of evolution is really consistent with a pattern where lineages are exploring new ecological space and changing rapidly to take advantage of that opportunity,” — Brian Weeks, also at the University of Michigan
