Comparing Chat GPT-4 Answers and Dictionary of Archival Terminology: Selected Examples

Abstract

This study aims to provide a clear comparison between ChatGPT-4's responses and the definitions found in the archival terminology dictionaries published by the International Council on Archives (ICA) and the Society of American Archivists (SAA), specifically addressing linguistic and conceptual challenges in archival science. Through comparative analysis of five selected archival terms from these dictionaries against ChatGPT-4's explanations, this research evaluates the accuracy, depth, and accessibility of AI-generated responses. The findings of this study are significant as they revealed the strengths and limitations of using generative artificial intelligence model ChatGPT-4 in the field of archival science. In addition, the results have shown that ChatGPT-4 demonstrates a notable ability to provide accessible and generally accurate explanations of archival terms. However, when comparing the depth of details and specificity to those found in the ICA and SAA dictionaries, it becomes evident that reliable traditional sources hold indispensable value for precise academic and professional purposes.