This article investigates the application of Machine Learning (ML), utilizing the Roboflow platform, to enhance the accuracy of identifying Glucophage 850 mg drug boxes, a widely used treatment for type 2 diabetes. The pharmaceutical industry faces persistent challenges, such as inventory discrepancies and drug authentication issues, which can lead to serious operational and patient safety concerns. To address these issues, this paper presents an advanced technological solution that leverages ML-driven image recognition techniques. We developed and tested a machine learning model capable of accurately detecting Glucophage 850 mg boxes. The model’s performance was assessed based on key metrics, including accuracy, reliability, and effectiveness, under real-world pharmacy conditions. Furthermore, a comprehensive literature review was conducted to outline recent advancements in object recognition technologies and their applications within the pharmaceutical sector, providing strong contextual support for our study. The results indicate that implementing such AI-based technologies could substantially reduce pharmaceutical distribution errors and improve inventory management. This demonstrates the potential of AI-driven solutions to revolutionize drug management practices, enhancing both operational efficiency and patient safety.