This project focuses on developing a robust, intelligent system for real-time pothole detection and driver alerts, using cutting-edge machine learning techniques, with a particular emphasis on the YOLO object detection model. The system, designed to improve road safety and minimize vehicle damage, continuously captures images from vehicle-mounted cameras, identifying and categorizing potholes in real-time. These potholes are classified into four levels of severity—Critical, High Risk, Moderate, and Low Risk—guiding drivers on the urgency and importance of avoiding each hazard. Operating within a 5-meter range, this system ensures that drivers receive timely warnings, allowing for informed decisions that reduce the risk of road accidents and vehicle wear. YOLO was chosen for its high-speed processing capabilities and accuracy, capable of rapidly analyzing road images and consistently detecting potholes even across diverse road and lighting conditions.  The integration of YOLO with a real-time notification system enables drivers to receive instant alerts, helping them navigate away from dangerous road areas, which promotes safer and more responsive driving. YOLO’s advanced computer vision features enable reliable performance and quick adaptation to changing environments, making the system versatile across urban, rural, and varied lighting conditions. Furthermore, continuous updates to the model improve detection accuracy, ensuring that the system remains efficient over time. This adaptability and reliability provide a proactive solution to road safety challenges associated with potholes. By increasing driver awareness, reducing vehicle maintenance costs, and encouraging safer driving practices, this project offers a comprehensive approach to tackling one of the most common hazards on the road, contributing to smoother, safer journeys for drivers everywhere.