The Role of ChatGPT in Enhancing Education: Evaluating Its Performance in Veterinary Physiology Assessments

Abstract

Artificial Intelligence (AI), particularly Generative Pre-trained Transformers (GPT) like OpenAI’s ChatGPT, has shown remarkable potential in revolutionising educational practices, primarily through its capacity to produce text that mimics human writing. This research evaluates ChatGPT's performance in addressing veterinary physiology exam questions, emphasising both subjective and objective question formats aligned with the MSVE-2016 curriculum. An observational cross-sectional study was carried out at the College of Veterinary Sciences, Rampura Phul, using ChatGPT 3.5 to respond to theory-based and multiple-choice (MCQ) exam papers. The AI demonstrated notable success with subjective questions, securing a 66% score. However, it encountered more difficulty with objective questions, achieving only 54%. These outcomes underscore the dual nature of ChatGPT's role as an educational aid in veterinary physiology—it offers promise but also reveals limitations. To maximise its effectiveness, it is recommended to apply it under the guidance of educators, ensuring accuracy and reliability. Additionally, there is a clear need to further refine the AI’s capabilities, especially in handling intricate, fact-based, and visual queries. Expanding AI training datasets with a specific focus on veterinary science will be crucial in enhancing its educational effectiveness moving forward.