Synthetic data in research: expert perceptions and a protocol for responsible GenAI use

Abstract

Purpose The purpose of this study is to assess the academic community’s perspectives and experiences regarding the academic, practical and ethical implications of using synthetic data generated by generative artificial intelligence (GenAI) tools in academic research. Design/methodology/approach The authors document and analyze perspectives from 32 distinguished hospitality and tourism scholars, including senior professors, journal editors and methodological specialists from around the world. Findings Hospitality and tourism scholars predominantly express a sentiment of caution regarding the use of synthetic data. Accordingly, they argue that GenAI-generated synthetic data cannot serve as a complete replacement for data obtained through traditional methods involving real human participants or other real-world applications. The use of synthetic data might be valuable as a supplementary tool, for example, in pilot testing, scale development and robustness tests, provided that it is used with complete transparency and strict human verification throughout these stages. Originality/value Drawing on expert consensus, they propose the GUARD Protocol as a perception-informed framework for responsible synthetic data use, intended as a starting point for community deliberation rather than a validated standard. This study makes an important contribution to the extant literature by establishing a comprehensive boundary condition (the GUARD Protocol) for the use of GenAI-generated synthetic data in academic research, hence moving the current debate beyond scattered commentary and synthesizing scholars’ perspectives into a unified framework.