Umeå universitets logga

Background: AI management has attracted increasing interest from researchers rooted in many disciplines, including information systems, strategy, and economics. In recent years, scholars with interests in these diverse fields have formulated similar research questions, investigated similar research contexts, and even often adopted similar methodologies when studying AI. Despite these commonalities, the AI management literature has largely evolved in an isolated fashion within specific fields, thereby impeding the development of cumulative knowledge. Moreover, views of AI’s anticipated trajectory have often oscillated between unjustifiably optimistic assessments of its benefits and extremely pessimistic appraisals of the risks it poses for organizations and society.

Method: To move beyond the polarized discussion, this work offers a systematic review of the vast, interdisciplinary AI management literature, based on analysis of a large sample of articles published between 2010 and 2022. Results: We identify four main research streams in the AI management literature and associated, conflicting discussion, concerning four (data, labor, critical, and value) dimensions.

Conclusion: The review conceptually and practically contributes to the IS field by documenting the literature’s evolution and highlighting avenues for future research trajectories. We believe that by outlining four key themes and visualizing them in an organized framework the study promotes a holistic and broader understanding of AI management research as a cross-disciplinary effort, for both researchers and practitioners, and provides suggestions that extend the framing of AI beyond myth and hype.