In a society at the forefront of technological change, autonomous vehicles (AVs) are emerging as a transformative mode of transportation. Despite their promise, they face significant challenges in perception, environment detection, and decision-making, particularly when detecting hazards like potholes. By leveraging advancements in artificial intelligence (AI), this research aims to address these limitations and enhance the safety of AVs and their passengers. The objective of this research is to develop a real-time pothole detection system using deep learning image-detection techniques. This system is designed to improve the reliability and accuracy of pothole detection, which remains challenging for AVs. Our current findings demonstrate the AI model's ability to identify potholes with significant confidence, indicating the feasibility and value of this system. With the use of our AI detection model, we can improve the safety and trustworthiness of AVs, thereby accelerating their mainstream adoption as a reliable mode of transportation.