Evaluating Human Expertise and Generative AI (ChatGPT) Responses to Questions on Water Challenges: A Quantitative Study

Suha Khalil Assayed () and Almoayied Assayed ()
Additional contact information
Suha Khalil Assayed: The British University in Dubai, Faculty of Engineering and IT
Almoayied Assayed: Royal Scientific Society, Water, Environment and Climate Change Centre

A chapter in Generative AI and Optimization Techniques for Sustainable Water Management, 2026, pp 13-33 from Springer

Abstract: Abstract This chapter presents a quantitative study investigating the differences between human expert and one of the most popular generative model (ChatGPT-4) in answering different of water-related questions with focusing on multiple evaluation criteria including clarity, depth, and flow. However, the participants answered to the questions without knowing their source, they don’t know if the answers are from human expert or ChatGPT in order to ensure the unbiased comparison. The Wilcoxon signed-rank test was applied to analyze whether perceived differences between experts and AI-generated responses were statistically significant. The findings reveal that there is a significant difference between the human experts and ChatGPT answers in all criteria, as the human experts provided more accurate and understandable answers. These results highlight the essential role of human expertise in technical domains where knowledge accuracy is critical. As well as the potential for generative AI to complement, rather than replace the human experts. In future work, a qualitative research should be conducted to explore more the participant perceptions and the impact of their background factors such as career position, age, education, and prior experience with water-related topics.

Keywords: Artificial intelligence; ChatGPT; Generative AI; Quantitative study; Water; Wilcoxon signed-rank test (search for similar items in EconPapers)
Date: 2026
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

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:spochp:978-3-032-19012-3_2

Ordering information: This item can be ordered from
http://www.springer.com/9783032190123

DOI: 10.1007/978-3-032-19012-3_2

Access Statistics for this chapter

More chapters in Springer Optimization and Its Applications from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().