RESPONSIBLE AI IN ORGANIZATIONS: EXPLORING MANAGERS’ PERCEPTIONS, CHALLENGES AND GOVERNANCE PRACTICE

Author: Malaysia

Abstract

Artificial intelligence (AI) is being adopted rapidly across organizations, yet the translation of high-level ethical principles into everyday managerial practice remains poorly understood. This study explores how managers perceive responsible AI (RAI), the challenges they face in implementing it, and the governance practices they adopt. Using a qualitative exploratory design, semi-structured interviews were conducted with 11 middle and senior managers from the banking, telecommunications, healthcare, public sector, higher education and manufacturing sectors in Malaysia. Data were analysed using reflexive thematic analysis supported by ATLAS.ti 24. Four themes emerged: (1) responsible AI as accountable and trustworthy decision-making, (2) the principle–practice gap, (3) emergent governance mechanisms, and (4) governance maturity as a continuum. Managers understood RAI primarily through the lenses of accountability, fairness, transparency and human oversight, and several anchored these in organizational and religious-ethical values such as amanah (trustworthiness). Key challenges included limited AI literacy, vendor opacity, regulatory ambiguity, resource constraints and unclear ownership. Governance practices were largely emergent, including cross-functional AI committees, risk-tiered approval, human-in-the-loop controls and vendor due diligence. The study contributes a manager-centred framework linking perceptions, challenges and governance maturity, and offers practical recommendations for organizations and policymakers seeking to operationalize responsible AI.