Nim is a classic type of game play, which rely on strategy as its main element. The game is played by two players with starting a series with the rod, where each player must break the series into two sets stems where the number of stems in each collection must not be the same and can not be empty. Nim game created with AI (Artificial Intelligence) specific. Utilization of these artificial intelligence systems for manufacturing applications include an intelligent game, this example is the game Nim. Talking about Artificial Intelligence or artificial intelligence, computer technology and the one evolving machine is one part of the science of informatics that has plenty of types of algorithms. There are many algorithms that can be used in the game of Nim, but that will be discussed in this thesis is the alpha-beta pruning algorithm. This algorithm is a modification of the Minimax algorithm. In theory the Alpha-beta pruning algorithm is superior compared to the Minimax algorithm, both from the ability to conduct the program in terms of searching and comparison of the time. Because the alpha-beta pruning algorithm can reduce the nodes in the algorithm Minimax search. This game has two important attributes in making a game, ie: number of sticks that would be played and who plays first. Keywords: NIM Game, Artificial Intelligence, Alpha-beta pruning, Minimax.