Now showing 1 - 2 of 2
  • Thumbnail Image
    Some of the metrics are blocked by your 
    Item type:Publication,
    Beyond control? Governing AI in organizations for responsible use
    Ensuring the responsible use of artificial intelligence (AI) has become a central concern for researchers, business leaders, and policymakers. There is a broad consensus that governing AI at the organizational level is crucial. While the existing literature acknowledges that responsible AI use depends on institutional structures and human agency, it is divided over which factor is decisive. We take an integrative perspective and examine how structure and agency interact when organizations govern AI for responsible use. To this end, we enrich Giddens' structuration theory with ethics (i.e., deontology, consequentialism, virtue ethics) and conceptualize AI governance as reflexive organizational praxis. We illustrate our analytical framework with the case of OTTO, Germany's largest e-commerce company. We conclude that while AI governance cannot define the responsible use of AI, organizations can scaffold how their members reason about whether and how AI should be used in a specific social context.
    Type:
    Journal:
  • Thumbnail Image
    Some of the metrics are blocked by your 
    Item type:Publication,
    How Ethical Theories Can Help Build Effective AI Governance in Organizations: Evidence From a Case Study at E-Commerce Company OTTO
    Governing artificial intelligence at the organizational level (AIG) is crucial to ensure that the technologies are used responsibly. The literature has largely focused on identifying suitable means to govern technologies that learn from data and are no longer deterministic. However, there remains considerable uncertainty about what ‘responsible’ use of AI means, and how governance will bring it about. This study addresses these conceptual issues by bringing ethical theories (consequentialism, deontology, and virtue ethics) into the AIG discourse. We first use these ethical theories to clarify the meaning(s) of responsibility and to develop action-theoretical mechanisms that relate governance to the organizational use of AI. Which we then apply in an in-depth case study at Germany’s leading e- commerce company, OTTO. We find that AIG encompasses three dimensions of responsibility (calculated, compliant, and competent use of AI) and that governance measures exhibit both the strengths and weaknesses of the ethical theories on which they are implicitly based. We conclude that governing AI for responsible use depends on situated judgment of individual decision makers but are currently under-theorized. Hence, we propose an alternative view that understands AIG as an ongoing, dialogical reflection between those who design AIG measures and those who enact them.
    Type: