Strategic Formation of CEO Apologies: Emulating Post-Crisis Public Statements Through GPT-4

Jeong-Ah Park () and Hye-Jin Kim ()
Additional contact information
Jeong-Ah Park: Korea Advanced Institute of Science and Technology
Hye-Jin Kim: Korea Advanced Institute of Science and Technology

Customer Needs and Solutions, 2024, vol. 11, issue 1, No 5, 17 pages

Abstract: Abstract In this research, we delve into the capacity of the GPT-4 artificial intelligence model to emulate CEO apology statements during crises, which are significant in shaping public trust and perception of a company. We introduce a novel method, “Prompt Engineering for CEO Apology,” which guides the AI to mimic distinct CEO writing styles and adapt to specific crisis contexts. Our study tests GPT-4’s capabilities using data beyond its last update in September 2021, assessing its adaptability to new and varied scenarios. Our proposed framework, grounded in the principles of rhetorical persuasion (ethos, pathos, and logos), incorporates detailed crisis and CEO writing style information into the prompts. We utilized the word2vec model and cosine similarity for a brief quantitative assessment of GPT-4’s ability to replicate CEO apologies, revealing its potential in mimicking human-like narratives. Additionally, we attempted to develop a specialized GPT model for CEO apologies, integrating our approach directly into its training for enhanced practical application. Our research aims to determine the most effective prompts for AI, marking a significant step toward using AI’s potential in human-like communication, with applications extending beyond business to politics, healthcare, education, and more.

Keywords: Prompt engineering; Apology statement; Artificial intelligence; Crisis communication; Content generation; ChatGPT (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s40547-023-00141-y Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:custns:v:11:y:2024:i:1:d:10.1007_s40547-023-00141-y

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/40547

DOI: 10.1007/s40547-023-00141-y

Access Statistics for this article

Customer Needs and Solutions is currently edited by Min Ding

More articles in Customer Needs and Solutions from Springer, Institute for Sustainable Innovation and Growth (iSIG)
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().