The use of artificial intelligence in a military context: development of the attitudes toward AI in defense (AAID) scale

Tools

Hadlington, L ORCID: https://orcid.org/0000-0001-9095-0517, Binder, J ORCID: https://orcid.org/0000-0002-1083-7109, Gardner, S ORCID: https://orcid.org/0000-0003-3443-7844, Karanika-Murray, M and Knight, S, 2023. The use of artificial intelligence in a military context: development of the attitudes toward AI in defense (AAID) scale. Frontiers in Psychology, 14: 1164810. ISSN 1664-1078

Preview

Text
1809764_Hadlington.pdf - Published version
Download (489kB) | Preview

Official URL: https://doi.org/10.3389/fpsyg.2023.1164810

Abstract

Introduction: The use of artificial intelligence (AI) for national defense is a matter of high societal significance and ongoing public discourse, but very little is known about public acceptance of AI in defense contexts. Currently, there is no reliable and valid measure of attitudes towards AI in defense, and more general attitudinal measures on AI use are unlikely to capture relevant perceptions and opinions. A measure was therefore developed for the assessment of Attitudes towards AI in Defense (AAID), and this work presents the initial validation of this scale.

Methods: A total of 1,590 participants (aged 19-75, M = 45.7, SD = 16.1) completed a self-report questionnaire which included an initial item pool of 29 attitudinal statements related to the use of AI in defense. An additional general attitude towards AI scale was also included to assess the concurrently validity of the AAID scale. The AAID underwent initial statistical validation via exploratory factor analysis and confirmatory factor analysis to test the underlying structure of the newly developed scale.

Results: Items reduction and exploratory factor analysis resulted in a final scale consisting of 15 items. A final two factor solution explained 42.52% of the variance (Factor 1 = 22.35%, Factor 2 = 20.17%). Factor 1 was termed “Positive outcomes” and reflected the potential and anticipated consequences of implementing AI in defense. Factor 2 was termed “Negative Outcomes” and reflected the potential negative outcomes for AI in defense. The scale also exhibited acceptable internal reliability and current validity.

Discussion: The newly developed AAID presents a new measurement tool which has the capacity to assess current attitudes towards AI in defense. Such work is essential if further developments in AI in defense are to continue with the support of the public. However, the work also notes that there are some key concerns and barriers that could block further developments in the area, with further work needed to explore how such anxieties are driven by narratives related to the topic.

Item Type:	Journal article
Publication Title:	Frontiers in Psychology
Creators:	Hadlington, L., Binder, J., Gardner, S., Karanika-Murray, M. and Knight, S.
Publisher:	Frontiers Media
Date:	5 May 2023
Volume:	14
ISSN:	1664-1078
Identifiers:	Number Type 10.3389/fpsyg.2023.1164810 DOI 1809764 Other
Rights:	© 2023 Hadlington, Binder, Gardner, Karanika-Murray and Knight. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Divisions:	Schools > School of Social Sciences
Record created by:	Laura Ward
Date Added:	28 Sep 2023 15:17
Last Modified:	28 Sep 2023 15:17
URI:	https://irep.ntu.ac.uk/id/eprint/49828