The influence of air pollution on human health is measured by the Air Quality Index (AQI). The AQI is a standardized scale that offers a monetary value to represent the degree of air quality in a specific location. The concentration of five main air pollutants—ground-level ozone, particle pollution, carbon monoxide, sulphur dioxide, and nitrogen dioxide—is used to compute the AQI. Prediction of the air quality index (AQI) can offer useful information that can assist people safeguard their health, schedule outside activities, and take actions to enhance air quality, simplifying and improving their lives. The aim of our research work is provide a comparative analysis of different artificial intelligence (AI) in prediction of air quality index (AQI). The AI model involved in our study are LSTM, BiLSTM, BiLSTMConv1D, and GRU. This research will help humans as well other livelihood, as significant number of livelihood experience illnesses brought on by air pollution each year, but only a small percentage of those endure fatalities. Overall based on our research proposal, we can conclude GRU out performs all other model and achieved the minimum values for Mean Squared Error (MSE)= 0.0006, Mean Absolute Error (MAE) = 0.01116 and Root Mean Squared Error(RMSE) = 0.0246.