Computer crimes are one of the important issues of criminal law, and therefore it is very important to deal with them. On the other hand, artificial intelligence has greatly changed the virtual space and technological innovations in the field of computers. The purpose of this article is to examine the question of how artificial intelligence is effective in dealing with computer crimes and improving cyber security performance. The findings indicate that considering human limitations and the fact that computer viruses and worms have become intelligent and flexible, the design of agents with intelligent sensors to deal with computer crimes can be considered. One of the fields that can be used to improve cyber security is artificial intelligence. Artificial intelligence is effective in dealing with computer crimes by helping to identify and detect crime and at the same time prevent it. Also, data security and network security through intrusion detection and botnet detection, e-mail security through spam detection and phishing attack detection, dealing with fake accounts, user data and information protection, authentication and detection of authentication fraud, program security and monitoring User activity is one of the most important measures of artificial intelligence in improving cyber security performance. Of course, artificial intelligence in improving cyber security faces some limitations, such as the lack of transparency in the decision-making process of artificial intelligence, the need for high data and the necessity of human participation and the interference of confusing variables.