Large Language Models (LLMs) can be used for tasks such as summarization, translation, text generation, and answering questions. This makes LLMs relevant to various aspects of academic work, including both research and teaching. This research uses the widely used GPT LLM to classify Decision Support System (DSS) papers into data-driven or model-driven systems. This analysis suggests the software is effective for this task, demonstrating good accuracy for examples that clearly fit into either of these classifications. The investigation’s results also revealed that LLM software encountered challenges, similar to human classifiers, with hybrid systems that could legitimately be categorized into both classes.