The rapid spread of counterfeit information on digital platforms poses significant challenges to media credibility and public trust. This research investigates the efficacy of various machine learning models, including Long Short-Term Memory, Recurrent Neural Network, and Bidirectional LSTM, for detecting fake news, particularly within the context of the COVID-19 pandemic. The research utilizes a comprehensive methodology encompassing data preprocessing, model training, and evaluation based on established performance metrics, including accuracy, precision, recall, and F1-score. A variety of models were assessed, including Logistic Regression, Support Vector Machine, Random Forest, Naive Bayes, K-Nearest Neighbors, Gradient Boosting, and XGBoost, with SVM demonstrating the highest accuracy of 82% and an F1-score of 86.7%. Through comparative analysis, this research elucidates the strengths and limitations of each model, underscoring the importance of integrating robust machine learning techniques in the fight against misinformation. The findings provide valuable insights for future research and practical applications in enhancing the reliability of news dissemination on social media.