What is it about?
Large Language Models (LLMs) are changing how software is built, and developers use them for very different purposes, such as information retrieval. Most studies on LLM adoption look at general trends, so we know little about how personal and behavioral factors relate to the use of LLMs for each specific purpose. In this study, we looked at the individual factors behind the choice to use LLMs for distinct software engineering purposes, based on an existing categorization of purposes by Khojah et al. We surveyed 188 software engineers and used a well-known model of technology acceptance (UTAUT2) together with a statistical technique (structural equation modeling) to study these relationships.
Featured Image
Photo by Milad Fakurian on Unsplash
Why is it important?
We found that the use of LLMs for each purpose depends on a different set of factors, and some factors can even reduce adoption when considered alone. This shows that LLM adoption in software engineering is more complex than general studies suggest. Based on these results, we give companies practical recommendations, for example on integrating LLMs smoothly into existing development environments and on sharing practices that improve information retrieval.
Perspectives
This paper continues our TOSEM study on cultural values and LLM adoption, using the same group of 188 software engineers. There we looked at LLM adoption in general; here we asked a more specific question: do the reasons change depending on what developers use LLMs for? They do, and I think this matters for companies, because one general strategy may not fit every use of LLMs.
Prof. Stefano Lambiase
Aalborg Universitet
Read the Original
This page is a summary of: Exploring individual factors in the adoption of LLMs for specific software engineering purposes, Information and Software Technology, December 2026, Elsevier,
DOI: 10.1016/j.infsof.2026.108295.
You can read the full text:
Contributors
The following have contributed to this page
