Artificial Intelligence, Ethics, and Organizational Trust: A Systematic Review and Future Research Agenda
TL;DR
A review of recent literature exploring AI capabilities and ethics and their relationship with organizational trust identifies gaps in current literature and suggests a future research agenda including multilevel trust, industry differences in AI and ethical considerations, and measurement of responsible AI.
Abstract
Artificial Intelligence (AI) systems are becoming deeply ingrained in organizational decision-making processes due to their ability to improve efficiencies, scalability and analytics-related accuracy. Consequently, organizations have faced more severe ethical issues around transparency and bias, as well as sustainability challenges around trustworthiness. While there has been some research conducted on AI capabilities, ethics and human trust in AI systems, research has remained siloed across academic fields. This paper aims to conduct a review of recent literature exploring AI capabilities and ethics and their relationship with organizational trust. The study will be completed by conducting a literature review on peer-reviewed articles published from 2020 to 2025 using a PRISMA-guided methodology. Articles will be reviewed based on keywords related to artificial intelligence-enabled decision-making, ethical issues in organizations, explainability and trust in organizational settings. After applying inclusion and exclusion criteria, four central themes were found: 1) AI as a source of competitive advantage for organizational decision-making; 2) AI ethics, including bias, opacity, and lack of accountability; 3) Trust as a multicomponent process developed through performance, transparency, and accountability; and 4) Responsible AI governance as a moderating factor between efficiency and legitimacy. There is existing research to suggest that trust in technology is formed not only by performance but by transparency and explainability factors as well as ethical acceptability. This paper offers several contributions. First, it combines both strategic and behavioral schools of thought into one framework for understanding how trust can be formed in AI-mediated decision-making processes. Second, it identifies gaps in current literature and suggests a future research agenda including multilevel trust, industry differences in AI and ethical considerations, and measurement of responsible AI. Overall, AI-enabled decision-making is not sustainable if the organization cannot effectively implement AI with ethical risk mitigation practices.
