How does AI perform as a tour guide? A user-based assessment through the ChatGPT tour guide performance model at Gordion

Abstract

Purpose Artificial intelligence (AI) is rapidly advancing and reshaping travel services. Despite its increasing employment as a tour guide, there is only limited identification of how AI performs in this role. To address this gap, this empirical study aims to identify ChatGPT’s competencies and shortcomings, as well as the dynamics shaping its tour guide performance. Design/methodology/approach This study adopts the Task-Technology Fit Theory and Technology Acceptance Model and conducts an experiment with 45 participants to explore ChatGPT’s performance as a tour guide at the UNESCO World Heritage site of Gordion, Türkiye. Findings The findings reveal that features such as image and voice assistants gain prominence, highlighting new competencies emerging through AI integration. However, as tour guiding requires extensive spatial mobility, external factors such as internet inaccessibility hinder the chatbot’s effectiveness. Additionally, ChatGPT is perceived as cost-effective compared to human tour guide services, making it a practical option for individual travel in heritage site contexts. Originality/value This paper is one of the first studies exploring AI’s performance in the tour-guiding context. Its novel contribution lies in explaining how AI performs as a tour guide by identifying its competencies and shortcomings, as well as revealing the facilitators and barriers associated with AI’s tour-guiding services. Ultimately, the proposed ChatGPT Tour Guide Performance Model contributes to the existing body of knowledge by examining the dynamics that shape ChatGPT’s tour guide performance.

Ähnliche Arbeiten