Date
Publisher
arXiv
Recent advances in Generative Artificial Intelligence (GenAI) have substantially improved the ability of large language models (LLMs) to generate and explain source code. However, their performance on authentic object-oriented programming (OOP) assessments remains insufficiently understood. This study evaluates five widely used GenAI systems, ChatGPT-5.2, DeepSeek-V3, Gemini 2.5 Flash, Claude Sonnet 4.5, and M365 Copilot, using programming tests and examination tasks from an introductory university OOP course. The generated solutions were assessed using the same grading criteria applied to students and compared with historical student results from the same course, as well as findings from the previous year. Common errors were also analyzed to identify recurring limitations across models. All evaluated GenAI systems achieved higher scores than the average student cohort and frequently obtained full marks on longer programming tasks. Nevertheless, they occasionally produced non-compiling code and continued to struggle with advanced OOP concepts, particularly interfaces, abstract classes, and certain inheritance-related tasks. Performance was also limited on graphics-related questions involving image interpretation. Compared with the previous year, the evaluated systems demonstrated noticeable improvements across most assessments while exhibiting several recurring error patterns. The findings provide an updated evaluation of the capabilities and limitations of contemporary GenAI systems on authentic introductory OOP assessments. They also offer evidence that can inform the design of programming assessments, the responsible integration of GenAI tools into software engineering education, and future studies evaluating the evolution of AI-assisted programming.
Study at a glance
- Question: How well do generative AI tools perform on introductory college computer programming tests compared to students?
- Setting and data: Programming tests, exam tasks, and historical student results from an introductory university object-oriented programming course.
- Method: Evaluates five AI systems on course programming tasks using student grading criteria and analyzes error patterns.
Why it matters
While generative AI has substantially improved at generating and explaining source code, how well these models perform on authentic object-oriented programming assessments remains insufficiently understood.
Key findings
- In an introductory university programming course, generative AI systems scored higher than the student average and frequently earned full marks on longer coding tasks.
- The AI models occasionally produced code that could not run and struggled with advanced programming concepts and tasks requiring image interpretation.
- Generative AI showed noticeable improvements across most assessments compared to the previous year, though several recurring error patterns persisted.
Summary drafted with AI from the paper’s abstract and may contain errors. See the full paper for details.
Who age?
Why use AI?
Study design
