Effects of Fairness and Explanation on Trust in Ethical AI

Angerschmid, A; Theuermann, K; Holzinger, A; Chen, F; Zhou, J

Effects of Fairness and Explanation on Trust in Ethical AI

Angerschmid, A Theuermann, K Holzinger, A Chen, F

Zhou, J

Permalink

Publisher:: SPRINGER INTERNATIONAL PUBLISHING AG
Publication Type:: Chapter
Citation:: Machine Learning and Knowledge Extraction, 2022, 13480 LNCS, pp. 51-67
Issue Date:: 2022-01-01

Embargoed

	Filename	Description	Size
	Trust_Explanation_fairness__CD_MAKE_2022_.pdf	Accepted version	770.49 kB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Embargoed
Open Access

This item is currently unavailable due to the publisher's embargo.

The embargo period expires on 11 Aug 2024

Full metadata record

Field	Value	Language
dc.contributor.author	Angerschmid, A
dc.contributor.author	Theuermann, K
dc.contributor.author	Holzinger, A
dc.contributor.author	Chen, F https://orcid.org/0000-0003-4971-8729
dc.contributor.author	Zhou, J https://orcid.org/0000-0001-6034-644X
dc.contributor.editor	Holzinger, A
dc.contributor.editor	Kieseberg, P
dc.contributor.editor	Tjoa, AM
dc.contributor.editor	Weippl, E
dc.date.accessioned	2022-12-24T04:35:29Z
dc.date.available	2022-12-24T04:35:29Z
dc.date.issued	2022-01-01
dc.identifier.citation	Machine Learning and Knowledge Extraction, 2022, 13480 LNCS, pp. 51-67
dc.identifier.isbn	9783031144622
dc.identifier.uri	http://hdl.handle.net/10453/164554
dc.description.abstract	AI ethics has been a much discussed topic in recent years. Fairness and explainability are two important ethical principles for trustworthy AI. In this paper, the impact of AI explainability and fairness on user trust in AI-assisted decisions is investigated. For this purpose, a user study was conducted simulating AI-assisted decision making in a health insurance scenario. The study results demonstrated that fairness only affects user trust when the fairness level is low, with a low fairness level reducing user trust. However, adding explanations helped users increase their trust in AI-assisted decision making. The results show that the use of AI explanations and fairness statements in AI applications is complex: we need to consider not only the type of explanations, but also the level of fairness introduced. This is a strong motivation for further work.
dc.language	en
dc.publisher	SPRINGER INTERNATIONAL PUBLISHING AG
dc.relation.ispartof	Machine Learning and Knowledge Extraction
dc.relation.ispartofseries	Lecture Notes in Computer Science
dc.relation.isbasedon	10.1007/978-3-031-14463-9_4
dc.rights	info:eu-repo/semantics/embargoedAccess
dc.subject.classification	Artificial Intelligence & Image Processing
dc.title	Effects of Fairness and Explanation on Trust in Ethical AI
dc.type	Chapter
utslib.citation.volume	13480 LNCS
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
utslib.copyright.status	embargoed	*
utslib.copyright.embargo	2024-08-11T00:00:00+1000Z
dc.date.updated	2022-12-24T04:35:27Z
pubs.publication-status	Published
pubs.volume	13480 LNCS

Abstract:

AI ethics has been a much discussed topic in recent years. Fairness and explainability are two important ethical principles for trustworthy AI. In this paper, the impact of AI explainability and fairness on user trust in AI-assisted decisions is investigated. For this purpose, a user study was conducted simulating AI-assisted decision making in a health insurance scenario. The study results demonstrated that fairness only affects user trust when the fairness level is low, with a low fairness level reducing user trust. However, adding explanations helped users increase their trust in AI-assisted decision making. The results show that the use of AI explanations and fairness statements in AI applications is complex: we need to consider not only the type of explanations, but also the level of fairness introduced. This is a strong motivation for further work.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/164554