It Takes a Village: A Distributed Training Model for AI-based Chatbots

TL;DR

This paper describes a collaborative, low-cost approach to developing an AI-powered chatbot in an academic library, emphasizing staff involvement for successful implementation and understanding labor requirements.

Contribution

It introduces a distributed training model involving library staff for developing AI chatbots, highlighting the importance of cross-organizational collaboration and labor recognition.

Findings

Staff involvement is crucial for chatbot success.

Leveraging existing data sources streamlines training.

Recognizing labor is essential for sustainable AI deployment.

Abstract

In Summer 2023, staff from the information technology and reference departments at the University of Delaware Library, Museums and Press came together in a unique partnership to pilot a low-cost AI-powered chatbot. The goal of the pilot is to learn more about student and faculty interest in engaging with this tool, and to better understand the labor required on the staff side. Reference librarians and other public facing staff, including student workers, were instrumental in helping to train the chatbot. This article discusses the development of prompts, leveraging of existing data sources for training materials, and workflows involved in the pilot. It argues that, when implementing AI-based tools in the academic library, involving staff from across the organization is essential to ensure buy-in and success. Although chatbots are designed to hide the effort of the people behind them, such labor can be substantial and needs to be recognized.