ChatGPT is an artificial intelligence (AI) interface where in when you feed a question, it attempts to provide human-like responses in terms of text. This is particularly useful for producing content, responding to the queries, and explaining the ideas. Henceforth, without much of user resistance, this technology has been embraced by the users. They use ChatGPT predominantly for academic purposes and acquiring information. Therefore, it becomes essential to evaluate the effectiveness of performance of ChatGPT which becomes the intent of this study. To assess the efficiency of ChatGPT, sentiment analysis is done on the ratings of its reviews. Various algorithms like BERT, Multinomial Naïve Bayes, XGBOOST classifier, and Random Forest classifier were deployed on the dataset that was obtained from Kaggle to perform sentiment analysis. These algorithms helped to cluster the ratings as a range where 4–5 was considered positive, 3 as neutral and 1–2 as negative. The model resulted in realizing the comprehending capabilities of ChatGPT with respect to the query raised by the user. The examining of user input, desired output, and the actual output helps us evaluate the quality of the understanding of the ChatGPT. The results of the study endorse that the responses or reviews that are rated and used for enhancing the performance of ChatGPT significantly and gradually shown progress on the subject. The scope of the study is to contribute towards the finetuning of the results regenerated by AI.