Distinguishing AI-generated text from human-written content is a critical challenge in modern natural language processing (NLP). This study leverages Logistic Regression (LR) as an efficient and interpretable model for text classification, outperforming alternatives like Random Forest and Support Vector Classifier in execution time and performance. Additionally, other models such as Naive Bayes, Shallow Neural Networks, Linear Regression, and Convolutional Neural Networks (CNN) were also experimented with, providing valuable insights into their capabilities and limitations for this task. Using TF-IDF vectorization to transform raw text into numerical features, the Logistic Regression model demonstrates superior reliability across multiple evaluation metrics. This work highlights the practical implications of text classification for applications such as authorship verification, content moderation, and ethical AI usage.