Objective In this paper, we investigate the impact that ChatGPT has on learner participation and learner autonomy within learning environments. We summarize student perspectives regarding benefits and limitations of ChatGPT, build prediction models for classifying such perspectives, and find dominant drivers for acceptance as well as concern. Method: We conducted a survey among 412 undergraduate students aged between 19 years to 21 years to gather data regarding usage of ChatGPT and its impact on learning. The data was preprocessed using cleaning, feature selection, and encoding methods. This paper explores how ChatGPT is affecting the way students learn, focusing on their engagement and independence. A survey of 412 undergraduate students was carried out to gather opinions on the benefits and drawbacks of using ChatGPT in academics. The responses were analyzed using standard preprocessing methods and machine learning models, including Logistic Regression, SVM, Random Forest, and XGBoost, to predict student perspectives. The results indicate that the majority of students believe that ChatGPT assists them in comprehending concepts better and aid in their learning. But some also raised concerns regarding plagiarism, dependence on the tool too much, and challenging recall of information. Out of the machine learning models experimented with, XGBoost performed the best, followed by Random Forest. Generally, the research indicates both advantages and pitfalls of employing ChatGPT in teaching and learning and calls for careful use and sound policies for ensuring that it comes to be adopted in a responsible manner.