Leveraging Microsoft Copilot (GPT-5) for Calculations and Interactive Data Visualization

Large language models (LLMs) are increasingly integrated into education, yet their ability to perform calculation-based tasks and generate scientific visualizations has been limited. This study evaluates Microsoft Copilot (GPT 5) for chemistry education across four domains: (1) chemical equilibrium, pH, titration, and buffer calculations; (2) data visualization using histograms, box plots, correlation plots, and heatmaps; (3) multivariate analysis of periodic table properties through principal component analy-sis (PCA); and (4) image interpretation and creation in classroom contexts. Thir-ty-three representative questions were tested without additional prompting. Copilot delivered accurate, step-by-step solutions for acid–base and equilibrium problems, generated high-quality visualizations directly from uploaded datasets, and produced PCA score and loading plots with proper data standardization. These results indicate that GPT 5 significantly improves over earlier LLM versions, offering a practical tool for enhancing conceptual understanding and data literacy in chemistry education. However, limitations persist in interpreting complex chemical imagery, requiring hu-man oversight. Future work should focus on refining multimodal accuracy and devel-oping pedagogical frameworks for responsible AI integration.