Indian roads are some of the most challenging to drive on, with high traffic volumes, diverse road conditions, and a wide range of vehicles. To address the safety concerns associated with driving on these roads; the detection of traffic signs and potholes can be done using YOLO algorithms. We can detect and classify an image using the various image classification or object detection models available. This chapter aims to conduct a comparative analysis between YOLOv5 and v8 to determine which yields the most accurate results. The YOLO model is a popular object detection model known for its accuracy and speed. The You Only Look Once (YOLO) algorithm can be used to give the driver precise and timely information about potholes and traffic signs, integrated into the advanced driver-assistance systems (ADAS) available in vehicles. The system’s development involves training YOLO algorithms with a large dataset of Indian roads and diverse traffic signs sourced from Kaggle and Roboflow. This initiative has the potential to mitigate accidents by alerting drivers to potential hazards and providing real-time information for informed decision-making while driving. The proposed research aims to enhance road safety in India. To be precise, we compare the effects of two YOLO algorithms, namely YOLOv5 and YOLOv8, and show that YOLOv5 outperforms YOLOv8 by taking suitable datasets and metrics for comparison.