Artificial Intelligence (AI) can play a significant role in the development of future smart cities. In order to provide better living, AI should be incorporated in daily activities of people and it can provide optimal solutions for various issues. Drowsiness while driving is a serious issue which needs to be addressed. Thus, effective driver drowsiness detection systems are needed. This chapter highlights the application of AI in drowsiness detection. In this chapter, two approaches have been utilized for drowsiness detection and their findings have been compared. In the first system, ensemble of regression trees has been trained using the extracted eyes landmarks to estimate the eyes landmark’s locations. Then, Eye Aspect Ratio (EAR) is computed and an alarm will alert the driver if the EAR was below the threshold value for a specific time. In the second approach, pre-trained dlib shape predictor has been utilized for predicting facial landmarks on an input image. Then, EAR and mouth aspect ratio (MAR) are computed and compared against a pre-defined threshold value for each of them. If the eyes remain closed for 15 consecutive frames, the driver will be altered. The presented systems have been tested and evaluated using test videos under various conditions. The first system achieved an accuracy of up to 98.6%. However, the second system achieved an accuracy of up to 98.7%. The experimental results show that the proposed approaches can be useful for monitoring the driver and give early warning.