PERSON
John Nash
The mathematician who proved that any situation of competing interests has a point of rest—and that the point of rest is not the same as the good outcome.
John Nash is the mathematician of the multi-agent moment. In two brief papers written before he was twenty-two, he gave the world the Nash equilibrium: a configuration of strategies, in any game among self-interested players, from which no single player has an incentive to deviate unilaterally. Under broad conditions such a point always exists, and its existence is the load-bearing result of modern
multi-agent AI. The trading algorithm, the bidding bot, the population of competing recommendation engines—each is searching, whether its designers use the word or not, for the fixed point Nash proved was always there. But Nash bundled his equilibrium with an unsettling corollary: a stable point is not necessarily a good point.
Coordination failure is not a malfunction of the agents; it is the equilibrium working exactly as the mathematics says equilibria work. And Nash himself, the man who formalized rationality, spent thirty years on the far side of it—consumed by paranoid schizophrenia before reasoning his way back by an act of will that presupposed