A Study of Generative Artificial Intelligence-Based Applications in Evaluating the Sketch Works of Undergraduate Art Students: A Comparison of ChatGPT and Claude

The general field of Artificial Intelligence (AI) in art education is different from other applications, but there is not much research that can be used as a benchmark for the reasons that AI is effective in artwork evaluation. The researcher studied the effectiveness of ChatGPT-4o and Claude-3.5-Sonnet AI language models for evaluating the artwork sketches created by undergraduate art students. This study helps to realise the impacts of Artificial Intelligence on education through its ability to generate tailored feedback and real-time teaching in art education. Evaluation results are adversely affected as they are evaluated using different methods. Though both techniques bring direct strengths to assessment, they must be blended in such a way as to provide creative and technical feedback into complete evaluation results for an AI assessment system. The development of the student's artistic practice would be better in the event of the receipt of combined advantages from AIs. Based on this, the research proposes to enhance the AI model to come up with innovative motivation for enhancing the technical support until an advanced AI assessment system is developed that will improve the quality of evaluation and artistic skills in pedagogical art, as well as coming up with innovative advancements.