Universal statistical signatures of evolution in artificial intelligence architectures
Summary
This paper analyzes 935 ablation experiments from 161 publications to show that AI architectural evolution follows the same statistical laws as biological evolution, including heavy-tailed fitness effect distributions and punctuated equilibria dynamics. The findings suggest that evolutionary statistical structure is substrate-independent, determined by fitness landscape topology rather than the mechanism of selection.
View Cached Full Text
Cached at: 04/21/26, 07:21 AM
Paper page - Universal statistical signatures of evolution in artificial intelligence architectures
Source: https://huggingface.co/papers/2604.10571
Abstract
The study finds that artificial intelligence architectural evolution follows the same statistical patterns as biological evolution, including similar fitness effect distributions and convergence dynamics.
We test whether artificial intelligence architectural evolution obeys the same statistical laws as biological evolution. Compiling 935 ablation experiments from 161 publications, we show that thedistribution of fitness effects(DFE) of architectural modifications follows a heavy-tailedStudent’s t-distributionwith proportions (68% deleterious, 19% neutral, 13% beneficial for major ablations, n=568) that place AI between compact viral genomes and simple eukaryotes. The DFE shape matches D. melanogaster (normalized KS=0.07) and S. cerevisiae (KS=0.09); the elevated beneficial fraction (13% vs. 1-6% in biology) quantifies the advantage of directed over blind search while preserving the distributional form.Architectural originationfollowslogistic dynamics(R^2=0.994) withpunctuated equilibriaandadaptive radiationinto domain niches. Fourteen architectural traits were independently invented 3-5 times, paralleling biological convergences. These results demonstrate that the statistical structure of evolution is substrate-independent, determined byfitness landscape topologyrather than the mechanism of selection.
View arXiv pageView PDFGitHub0Add to collection
Get this paper in your agent:
hf papers read 2604\.10571
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2604.10571 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2604.10571 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2604.10571 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
Why Specialization Is Inevitable
This article argues that specialization is inevitable for AI systems, drawing on evidence from optimization theory, evolutionary biology, competitive markets, and machine learning. It interprets a 2026 paper by Goldfeder, Wyder, LeCun, and Shwartz-Ziv to challenge the assumption that greater capability leads to greater generality.
From Application-Layer Simulation to Native Meta-Architecture: Structural Tension as an Endogenous Driver for Heterogeneous AI Evolution
This paper proposes a theoretical framework for embedding higher-order cognitive architectures natively into LLMs via structural tension, offline recurrent loops, and inference-time plasticity, aiming for heterogeneous AI evolution under hard governance constraints.
Biological evolution and information acquisition
This article draws parallels between biological evolution and technological evolution, explaining how modularity and sexual reproduction allow populations to increase the rate of information acquisition. Simulations demonstrate that mixing genetic material accelerates the spread of beneficial mutations, analogous to how technologies build on existing components.
Uneven Evolution of Cognition Across Generations of Generative AI Models
This paper introduces a psychometric framework and the AIQ Benchmark to evaluate the cognitive profiles of generative AI models, revealing uneven evolution with strong verbal skills but stagnant perceptual reasoning.
Neuroevolution Arena: Nested Ecological Evaluation of Update-and-Inheritance Regimes across Neural Architectures
This preprint introduces Neuroevolution Arena, a GPU-accelerated artificial life platform for comparing evolution and RL-based update regimes across neural architectures via a nested ecological evaluation protocol.