Diversity and AI-Assisted Decision-Making:Can Diversity Help Reduce Bias?

Xiaoya Zhou

doi:10.47297/taposatWSP2633-456913.20260707

PDF
导出
分享
收藏
专辑

Diversity and AI-Assisted Decision-Making:Can Diversity Help Reduce Bias?

7卷第7期

浏览量: 0 Downloads: 0

Diversity and AI-Assisted Decision-Making:Can Diversity Help Reduce Bias?
Diversity and AI-Assisted Decision-Making:Can Diversity Help Reduce Bias?
Queen Mary University of London, Suzhou, Jiangsu, 215028, China
2026年7卷第7期页码：71-75
网络首发：2026-08-21，

纸质出版：2026-07-30
DOI：10.47297/taposatWSP2633-456913.20260707
Accepted：

Scan QR Code

Xiaoya Zhou. (2026). Diversity and AI-Assisted Decision-Making:Can Diversity Help Reduce Bias?. 科技理论与实践（英文版）, 7(7), 71-75.

Zhou Xiaoya.Diversity and AI-Assisted Decision-Making:Can Diversity Help Reduce Bias?[J].Theory and Practice of Science and Technology,2026,07(07):71-75.
Xiaoya Zhou. (2026). Diversity and AI-Assisted Decision-Making:Can Diversity Help Reduce Bias?. 科技理论与实践（英文版）, 7(7), 71-75. DOI： 10.47297/taposatWSP2633-456913.20260707.

Zhou Xiaoya.Diversity and AI-Assisted Decision-Making:Can Diversity Help Reduce Bias?[J].Theory and Practice of Science and Technology,2026,07(07):71-75. DOI： 10.47297/taposatWSP2633-456913.20260707.

摘要

Abstract

Artificial intelligence is increasingly used to support decisions in recruitment

healthcare

finance and public services. Although these systems are often presented as neutral and efficient

they can reproduce unfair patterns when training data

target variables and review processes are weak. This paper asks whether diversity can help reduce bias in AI-assisted decision-making. It argues that diversity is not a simple technical fix

but it can reduce bias indirectly when diverse participants have enough information and authority to influence design

testing and deployment. The paper uses a concise literature review

two case studies and an illustrative secondary-data analysis based on a preliminary student presentation dataset of 60 observations. The empirical section reports a preliminary logistic regression in which a higher diversity index is associated with a lower probability of a reported AI bias event. This result is treated as exploratory rather than causal. The Amazon recruitment AI case shows how historical data and weak review can reproduce gender bias

while the Google diabetic-retinopathy AI case shows how professional and local participation can improve deployment review in healthcare. Overall

the paper finds that diversity helps mainly by making bias easier to detect

making AI outputs easier to question

and making correction more likely through governance and feedback.

关键词

Keywords

references

OECD. (2024). OECD AI Principles. OECD.AI Policy Observatory.

Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K. and Galstyan, A. (2021). A Survey on Bias and Fairness in Machine Learning. ACM Computing Surveys.

Suresh, H. and Guttag, J. (2021). A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cycle. EAAMO.

Parasuraman, R. and Riley, V. (1997). Humans and Automation: Use, Misuse, Disuse, Abuse. Human Factors.

Hong, L. and Page, S. E. (2004). Groups of Diverse Problem Solvers Can Outperform Groups of High-Ability Problem Solvers. PNAS.

Phillips, K. W., Liljenquist, K. A. and Neale, M. A. (2009). Is the Pain Worth the Gain? The Advantages and Liabilities of Agreeing with Socially Distinct Newcomers. Personality and Social Psychology Bulletin.

National Institute of Standards and Technology. (2023). Artificial Intelligence Risk Management Framework (AI RMF 1.0). U.S. Department of Commerce.

European Parliament and Council of the European Union. (2024). Artificial Intelligence Act. Official Journal of the European Union.

Mitchell, M. et al. (2019). Model Cards for Model Reporting. Proceedings of the Conference on Fairness, Accountability, and Transparency.

Gebru, T. et al. (2021). Datasheets for Datasets. Communications of the ACM.

Raji, I. D. et al. (2020). Closing the AI Accountability Gap: Defining an End-to-End Framework for Internal Algorithmic Auditing. FAT*.

Dastin, J. (2018). Amazon Scraps Secret AI Recruiting Tool that Showed Bias against Women. Reuters.

Beede, E. et al. (2020). A Human-Centered Evaluation of a Deep Learning System Deployed in Clinics for the Detection of Diabetic Retinopathy. CHI.

Ruamviboonsuk, P. et al. (2022). Real-Time Diabetic Retinopathy Screening by Deep Learning in a Multisite National Screening Programme. The Lancet Digital Health.

Zhou, X. (2025). Diversity and AI-Assisted Decision-Making: Can Diversity Reduce Bias? Research Proposal Presentation and Preliminary Dataset, n = 60. Unpublished student presentation.

McKinsey & Company. (2023). Diversity Matters Even More: The Case for Holistic Impact.

Gartner. (2024). Gartner Survey Shows AI-Related Risks See Greatest Audit Coverage Increases in 2024.