Why users trust generative AI: an empirical exploration of trustworthiness dimensions and privacy disclosure
- 8 June 2026
- journal article
- Published by Emerald Publishing in Online Information Review
- Vol. 50 (4) , 837-854
Abstract
Purpose: As generative AI technologies increasingly mediate user interactions and services, trust in these systems has become a key determinant of user engagement and data-sharing behavior. This study investigates trust formation mechanisms driving users’ willingness to disclose personal information to generative AI, identifying key trustworthiness dimensions and examining their impact on perceived trust and disclosure intention. Design/methodology/approach: Two empirical studies were conducted. Study 1 employed exploratory and confirmatory factor analyses (N = 557) to develop a multidimensional trustworthiness scale, while Study 2 applied structural equation modeling (N = 251) to assess how these factors affect trust in AI and subsequent disclosure intention. Findings: Six trustworthiness factors were identified under competence (intelligence, company reputation and technical security) and warmth (psychological security, responsibility and comfort) dimensions. Psychological security, responsibility and technical security significantly predicted users’ trust in generative AI, which strongly influenced intention to disclose personal information. Originality/value: By focusing on the multidimensional nature of AI trustworthiness, this study identifies key antecedents and psychological mechanisms underlying users’ trust in generative AI and disclosure behavior. It contributes to the literature by developing and validating a reliable scale for measuring the trustworthiness of generative AI systems. The findings highlight that trust is shaped not only by technical factors such as technical security but also by perceived moral dimensions, including psychological security and responsibility, offering practical implications for designing AI systems that are both functionally robust and ethically reassuring. Peer review: The peer review history for this article is available at: https://publons.com/publon/10.1108/OIR-07-2025-0550Keywords
This publication has 53 references indexed in Scilit:
- Security and Privacy Concerns in ChatGPT2023
- Generative AIBusiness & Information Systems Engineering, 2023
- User trust in artificial intelligence: A comprehensive conceptual frameworkElectronic Markets, 2022
- Privacy in smart speakers: A systematic literature reviewSecurity and Privacy, 2022
- ‘Okay google, what about my privacy?’: User's privacy perceptions and acceptance of voice based digital assistantsComputers in Human Behavior, 2021
- Alexa, she's not human but… Unveiling the drivers of consumers' trust in voice‐based artificial intelligencePsychology & Marketing, 2021
- A multidimensional conception and measure of human-robot trust2021
- Trust data collections via vehicles joint with unmanned aerial vehicles in the smart Internet of ThingsTransactions on Emerging Telecommunications Technologies, 2020
- Promoting Trust Between Patients and Physicians in the Era of Artificial IntelligenceJAMA, 2019
- A new criterion for assessing discriminant validity in variance-based structural equation modelingJournal of the Academy of Marketing Science, 2014