Prisoner's dilemma
Two rational players each defect — and both end up worse off than if they had cooperated.
What it means
The prisoner's dilemma is the canonical game of cooperation in which each of two players independently chooses to cooperate or defect, with payoffs arranged so that mutual cooperation is better for both than mutual defection, yet each player earns more by defecting regardless of what the other does. Because defection is the dominant strategy for each individual, rational self-interest drives both to defect, producing a collectively bad outcome — the central paradox is that individually rational choices yield a jointly irrational result, with mutual defection as the only Nash equilibrium. The mechanism is the gap between individual and collective incentives: the temptation to defect on a cooperator, and the fear of being the exploited cooperator, both push toward defection. The picture changes profoundly with repetition: in the iterated prisoner's dilemma, the shadow of future interaction allows reciprocity-based strategies such as tit-for-tat to sustain cooperation through the threat of retaliation, and reputation has a similar effect. It matters because the dilemma models an enormous range of real conflicts — arms races, price wars, climate negotiation, trust between firms — where everyone would gain from cooperation but each is tempted to defect.
Examples
Two suspects each betray the other to cut their own sentence, landing both with worse prison terms than if they had both stayed silent.
Two petrol stations on the same road would both do better holding prices up, but each can win the week by undercutting — so both cut, and both earn less.
Two flatmates each leave the washing-up, reasoning the other might crack first. Neither gains from the standoff, yet scrubbing while the other free-rides feels worse than a dirty kitchen.
First described in Flood & Dresher (1950); framed by Albert Tucker.