You On AI Field Guide · The Evolutionarily Stable Strategy The You On AI Field Guide Home
TxtLowMedHigh
CONCEPT

The Evolutionarily Stable Strategy

John Maynard Smith's central concept: a way of behaving that, once common in a population, cannot be displaced by any invading alternative—order without a referee, and the precise mathematical object that multi-agent AI systems converge toward by a completely independent route.
The evolutionarily stable strategy is the answer to a question that troubled evolutionary biologists for decades: why do populations of self-interested agents settle into the patterns they do, without any authority directing them? John Maynard Smith, with George Price, formulated the concept in 1973 as a refinement of the Nash equilibrium suited to populations of agents that adjust through selection rather than rational deliberation. A strategy is evolutionarily stable when, once nearly everyone in the population plays it, no mutant strategy introduced in small numbers can achieve a higher payoff and spread. Stability is defined against invasion: not merely that no one wants to deviate right now, but that any deviation is punished and fails to spread. This is the engineer's move in biological form—the difference between a structure that is balanced and one that returns to balance when nudged. The concept arrives in the AI era bearing a structural surprise: researchers
← Home0%
CONCEPTBook →

Keep reading with YOU ON AI

Unlock the full book, 10,000+ field-guide entries, and a 1000+ thinker library. If you have a book code, register now — it takes a minute.

Register with book code Sign in