In Artificial Intelligence (AI) We (Dis)Trust? Navigating Institutional Pressures for Automation and Augmentation in the Implementation of AI in Organizations

Giannitsas, Dimitris and Sun, Ruonan and Baptista, Joao (2026) In Artificial Intelligence (AI) We (Dis)Trust? Navigating Institutional Pressures for Automation and Augmentation in the Implementation of AI in Organizations. Information and Organization. ISSN 1471-7727 (In Press)

Text (I&O_Manuscript_Round 4_Final)
I_O_Manuscript_Round_4_Final.pdf - Accepted Version
Available under License Creative Commons Attribution.
Download (926kB)

Abstract

AI brings competing demands to organizations from pressures towards efficiency and standardization versus contextual responsiveness and ethical judgment. These demands become particularly salient when some areas of organizations push for automation while others for augmentation, as two distinct paradigms of AI implementation. It is therefore important to understand how organizations navigate these competing institutional pressures. Drawing on a nested case study of a European airline, we find that the choice between automation and augmentation is not solely a matter of task–technology fit. It is instead also shaped by how actors configure trust and distrust in AI systems in response to two coexisting institutional logics: instrumental–analytic and contextual–normative. We show how these two logics stimulate different trust–distrust configurations, which in turn guide how AI is implemented and adopted within organizations. We identify two reconciliation practices that help organizational actors manage inherent tensions between these competing institutional pressures: mindful evaluation and proactive safeguarding. The research reveals how AI implementation and adoption reflects conflicts between dominant institutional logics and contributes with a novel perspective on the role of institutional logics and trust in projects of AI implementation.