Can ChatGPT code the technical part of a Bachelor's Thesis in Informatics?

Abstract

This study delves into the potential of Large Language Models (LLMs), specifically ChatGPT by OpenAI, in assisting with computer code writing for Bachelor's theses in the field of Information Systems at Dalarna University, Sweden. While LLMs have shown promise in various applications, their efficacy in academic coding tasks remains underexplored. Through a pilot study, we investigated the extent to which ChatGPT can support or even take over the coding aspect of a Bachelor's thesis. Our findings indicate that ChatGPT can expedite the coding process and empower students to undertake technical analyses. However, the success of this collaboration is contingent on the student's ability to engage in a "critical dialog" with the model, assessing its outputs and seeking refinements as needed. While our results are context-specific and tied to the competencies of the involved students, they underscore the potential of LLMs in academic coding tasks and highlight the need for broader investigations in this domain.