Generative Artificial Intelligence and Studying Programming Subjects at Universities: From Students Motivation to Comparative Analysis of AI and Human Code

Abstract

The rapid expansion of generative artificial intelligence (AI) is reshaping process of studying programming subjects, as students increasingly employ AI tools to solve academic problems. In this paper we investigate student’s motivation for using generative AI, their perceived advantages and drawbacks, and the distinctions between human-written and AI-generated code. Building upon earlier research on the impact of AI on academic performance, this paper analyses an experimental dataset comprising
950 code samples derived from student and AI solutions to identical programming tasks. The dataset was processed using Python-based analytical tools to extract structural and stylistic metrics, enabling a comparative assessment of the behaviour of different generative models. The findings reveal that certain models, particularly ChatGPT, exhibit greater stylistic proximity to student code, while others demonstrate more formalized programming patterns. The results contribute to a deeper understanding of AI’s
role in process of studying programming subjects and provide a methodological foundation for developing tools to evaluate and monitor AI-assisted learning.