Cosmological Natural Selection

In the AI Story

The fine-tuning problem has haunted physics since the mid-twentieth century. The physical constants — the strength of gravity, the mass of the electron, the cosmological constant, the coupling constants of the fundamental forces — appear to be exquisitely tuned for the production of complex structures. Change any of them by a small amount, and stars do not form, or atoms do not bind, or the universe recollapses before chemistry has time to develop. The question of why the constants take the values they do has no satisfying answer within standard physics. The constants are treated as brute facts, fixed parameters that must be measured but cannot be derived.

Smolin's hypothesis treats the constants as variables subject to evolutionary pressure. The mechanism is the quantum gravitational process that occurs at the singularity of a black hole — or rather, in the region that replaces the classical singularity in any quantum theory of gravity. This process, on Smolin's proposal, produces a new region of spacetime: a baby universe, causally disconnected from its parent, with its own big bang and its own physics. The physical constants of the baby universe are inherited from the parent with slight variations — mutations, in the biological analogy — produced by the quantum gravitational dynamics.

The selection pressure is straightforward. Universes whose constants favor the formation of many black holes produce many baby universes, which themselves tend to have constants favoring black hole formation, which produce many more baby universes. Over cosmological time, the multiverse becomes dominated by constants optimized for black hole production. The constants we observe in our universe should therefore be near a local maximum of black hole production — a testable prediction that distinguishes cosmological natural selection from anthropic reasoning.

The remarkable feature of the hypothesis is that the constants optimized for black hole production are also the constants that produce complexity. Black holes form from the collapse of massive stars, which require nuclear physics capable of producing heavy elements. Heavy elements require electromagnetism strong enough to bind complex molecules. Complex molecules, given sufficient time and energy flow, produce the self-organizing chemistry that Kauffman describes. Life, consciousness, and technology are downstream consequences of the same constants that maximize black hole production. The universe did not fine-tune itself for us. It fine-tuned itself for black holes. We are a predictable side effect — but a predictable one.

For the AI discourse, the implications are substantial. If cosmological natural selection is correct, the river of intelligence is not an accident but an expression of cosmological selection operating across generations of universes. The emergence of AI is not a technological aberration but the latest channel through which a selected tendency toward complexity finds expression. The framework does not make AI inevitable in any specific form — contingency operates at the level of specifics — but it makes the general direction toward increasing complexity a feature of the universe rather than a coincidence.

Origin

Smolin proposed cosmological natural selection in a 1992 paper and developed it at length in The Life of the Cosmos (1997). The hypothesis has been discussed by Stuart Kauffman, Roger Penrose, and others across the intersection of cosmology, quantum gravity, and the philosophy of science. It has not been decisively confirmed or refuted, but it remains one of the few hypotheses that offers a non-anthropic explanation for fine-tuning.

Key Ideas

Universes reproduce. Black holes produce new regions of spacetime with slightly varied physical constants — a cosmological analog of biological reproduction.

Selection operates on constants. Universes whose constants favor black hole production come to dominate the multiverse over cosmological time.

Complexity as side effect. Constants optimized for black holes also produce stars, heavy elements, chemistry, and life — not by design but as a predictable correlation.

Fine-tuning explained. The apparent tuning of constants for complexity is the expected outcome of cosmological selection, not a mystery requiring anthropic or multiverse reasoning.

Testable. The hypothesis predicts that our universe's constants should be near a local maximum of black hole production — a testable claim that distinguishes it from non-explanatory alternatives.

Debates & Critiques

The hypothesis has faced technical challenges regarding the specific mechanism by which baby universes inherit modified constants and whether the variation is small enough to permit cumulative selection. Critics argue that without a worked-out theory of quantum gravity, the reproductive mechanism is speculative. Supporters respond that the hypothesis is empirically testable in principle and offers the only non-anthropic explanation currently on offer.