In recent years, there has been an increase in traffic accidents caused by distracted driving. According to a 2022 report by the Federal Ministry of Road Transport and Highways, in India have 18 deaths in one hour in two road accidents. Therefore, measures to reduce traffic fatalities are essential. Tragically, the number of lives lost on the roads this year has reached a peak not seen since 2014, with 155,622 people losing their lives in preventable accidents. This is a stark reminder of the importance of safe driving practices and the need to prioritize road safety in our communities. The main cause of these accidents is driver error. This document proposes solutions for detecting driver distraction and avoiding potential accidents. In this document, we present the use of various convolution neural network (CNN) models for classifying distracted drivers. These models include small-scale CNN, VGG16, VGG19, and the Kaggle State Farm Distracted Driver Detection Challenge model. We are using the Keras library with Tensor Flow as our deep learning platform. Our highest performing model achieved a categorical cross-entropy loss of 0.899 on the validation set. These models may serve as a starting point for further research on distracted driver detection.