Driver fatigue and aggressive driving behavior are among the major causes of road accidents across the world, creating significant challenges for transportation safety. Traditional driver monitoring approaches generally depend on hardware-based sensors and manual observation methods, which may involve higher costs and limited efficiency in real-time analysis. This paper presents an AI-Based Driver Drowsiness and Road Rage Prediction System that utilizes intelligent behavioral analysis for identifying unsafe driving patterns and enhancing road safety measures. The proposed model evaluates multiple driver-related factors, including driving time, steering movement patterns, fluctuations in vehicle speed, braking activities, stress-related indicators, and driver behavioral responses to detect symptoms of drowsiness and road rage. Machine learning techniques such as Random Forest, Logistic Regression, and XGBoost are applied for behavior classification and predictive analysis. Experimental results indicate that the system achieves higher prediction performance and effective risk assessment under various driving scenarios. The developed framework can contribute to intelligent transportation systems by enabling proactive monitoring of driver behavior and minimizing accident possibilities