Impact of AI Debugging on Student Code Complexity and Readability: A Literature Review

Debugging is a critical yet challenging skill for novice programmers. This literature review examines how Al-assisted debugging influences the complexity and readability of student-written code, compared to traditional human-guided debugging. We present a historical overview of instructor-led debugging pedagogy and contrast it with emerging Al debugging assistants. We then analyze the general impacts of Al-assisted debugging tools on coding practices, focusing on qualitative factors such as code maintainability, clarity, and student perceptions. The review spans various educational contexts, from university courses to high school classes and coding bootcamps, to assess how different learner populations are affected. We find that Al assistants can accelerate the debugging process and sometimes produce code of similar complexity and readability to human-developed code. However, concerns remain about overreliance, superficial understanding, and maintaining code clarity. Comparisons with traditional debugging highlight trade-offs between efficiency and learning depth. This review sheds light on the current findings and offers an analytical perspective on integrating Al-assisted debugging into programming education.