This paper argues that a precise and defensible version of the nineteenth-century idea that "ontogeny recapitulates phylogeny" re-emerges through the lens of modern generative models. The species genome is framed not as a record of ancestral forms, but as a compressed, principal-component encoding of survival knowledge—a generative model trained by evolution's variation-and-selection over deep time. Development (ontogeny) is modeled as the generative reverse pass (a diffusion cycle) where each organism is regenerated as a fresh, plausible variant and verified in real time against its context. The paper integrates the transcriptomic developmental hourglass, the formal correspondence between evolution and learning, and information-theoretic constraints.
No takes yet. Share an insight, caveat, or question.
Bogdan Motoc (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: