In mobile security and privacy, android malware detection represents an important security challenge in the Android operating system. Since Android is customizable and an open-source operating system, malware such as spyware, Trojans, keyloggers, and other forms of malware can be injected by attackers to penetrate Android-based mobile devices. In response to these security threats, this paper introduces a practical comparison study to investigate an artificial intelligence framework consisting of nine machine learning techniques, and two neural networks techniques for accurately detecting android malware. For testing the efficiency of the proposed AI framework in detecting Android malware, a benchmark and labeled malware dataset from the Kaggle platform has been used to train and test the proposed AI framework. The experimental results showed that Random Forest and Gradient Boosting machines (GBM) outperformed other machine and Neural networks models in detecting Android malware, with malware detection accuracy reaching 97%. These findings can support strong recommendations to develop more robust and secure AI models for detecting recent Android attacks such as BingoMod malware, one of the recent Remote access trojans (RATs) android malware that allows cybercriminals to make unauthorized money transfers remotely.