AUTHOR=Almuraqab Nasser A. Saif , Ateeq Ali , Manoj Kumar M. V. , Alfiras Mohanad 

TITLE=Public attitudes toward secure AI enabled drone delivery for public services in the UAE

JOURNAL=Frontiers in Built Environment

VOLUME=Volume 11 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/built-environment/articles/10.3389/fbuil.2025.1640830

DOI=10.3389/fbuil.2025.1640830

ISSN=2297-3362

ABSTRACT=IntroductionSecure artificial intelligence (AI)-enabled drone delivery systems are emerging as transformative solutions for public service delivery, particularly in smart governance contexts such as the United Arab Emirates (UAE). While promising, the adoption of these systems requires a nuanced understanding of factors influencing public acceptance, including AI-security assurances, perceived risks, costs, and social influence.MethodsThis study uses a survey of 410 UAE residents, analyzed through partial least squares structural equation modeling (PLS-SEM), to explore the drivers of public acceptance of AI-enabled drone systems. The study integrates these factors into a structural acceptance model, focusing on the roles of perceived benefits, risks, costs, and social influence.ResultsThe findings demonstrate that both perceived benefits (β=0.386, p<0.001) and social influence (β=0.386, p<0.001) are strong and significant drivers of positive attitudes towards AI-enabled drone delivery systems. In contrast, perceived risks negatively impact acceptance (β=−0.146, p=0.002). Interestingly, perceived cost does not significantly affect attitudes (β=−0.057, p=0.445) but is positively associated with risk perceptions, indicating a layered barrier effect.DiscussionThe study contributes to technology acceptance models by revealing the interdependencies between barrier constructs. It suggests that in the UAE context, public engagement and security assurances are more crucial for fostering trust and adoption than cost-related incentives. Limitations of the study include nonrandom sampling, a cross-sectional design, and weaker loadings for certain indicators, which may limit generalizability but provide valuable exploratory insights.