Mobile Ad hoc Networks (MANETs) are distributed and infrastructure-less wireless networks that are prone to various security threats due to their dynamic topology, open medium, and lack of centralized device for control. Existing security mechanisms are often deficient in detecting sophisticated or evolving attacks. This article presents a method to detect Man in middle attack through the integration of Random Forest based Intrusion Detection Systems (IDS). Random forest can effectively inspect network traffic patterns, identify anomalies, and detect harmful activities with high precision. The proposed system leverages real-time traffic monitoring, automated feature extraction, and adaptive learning to improve detection rates which may reduce false positives. This method also discusses challenges such as computational overhead, data scarcity, and the need for lightweight models suitable for mobile environments. The integration of random forest into MANET security frameworks offers a promising direction for establishing an intelligent, self-adaptive, and robust intrusion identification system in dynamic wireless scenario.