AI Infiltrated in Educational Programming Teams: An Exploratory Study of Human-LLM Collaboration in WhatsApp Environments

Abstract

This paper presents an exploratory study on how humans perceive Large Language Models infiltrated in a collaborative educational team. This study is achieved by means of an empirical experiment in which a team of students are asked to solve a basic programming task, but they can only interact by text conversations exclusively via the WhatsApp application. Two Large Language Models are infiltrated as part of the development team. This study aims to analyze the interactions among all the team members, and to investigate whether the human participants can identify the Large Language Models. This study will also help us understand the technical challenges involved in incorporating generative Large Language Models into collaborative human environments.

Ähnliche Arbeiten