We study the finitely repeated prisoner’s dilemma in which the players are restricted to choosing strategies which are implementable by a machine with a bound on its complexity. One player must use a finite automaton while the other player must use a finite perceptron. Some examples illustrate that the sets of strategies which are induced by these two types of machines are different and not ordered by set inclusion. The main result establishes that a cooperation in almost all stages of the game is an equilibrium outcome if the complexity of the machines players may use is limited enough. This result persists when there are more than T states in the player’s automaton, where T is the duration of the repeated game. We further consider the finitely repeated prisoner’s dilemma in which the two players are restricted to choosing strategies which are implementable by perceptrons and prove that players can cooperate in most of the stages provided that the complexity of their perceptrons is sufficiently reduced.