Author Information : Natarajan Balasubramanian (Whitman School of Management, Syracuse University)
Yang Ye (Southwestern University of Finance and Economics - Chengdu, China)
Mingtao Xu (Louisiana State University)
Year of Publication : Academy of Management Review (Forthcoming)
Summary of Findings : This paper provides a theoretical investigation of how the substitution of human decision-making with machine learning algorithms may influence organizational learning.
Research Questions : a) Can substituting human decision-making with machine learning algorithms possibly impoverish organizational learning?
b) What, if any, are some important factors that may amplify or mute this effect?
What we know : Over the past few years, machine learning algorithms have been adopted by many types of organizations, including large and small businesses, hospitals, police, and schools, not all of which may have the requisite resources and capabilities to fully understand the risks of substituting human decision-making with machine learning. Extant literature suggests that the adoption of machine learning benefits organizations by increasing the speed and volume of information processing. Notwithstanding these benefits, machine learning has some inherent limitations due to its reliance on statistical analysis of historical data. We do not know if and how these limitations may affect organizational learning.
Novel Findings : a) To our knowledge, we are the first study to systematically investigate and lay out the possible impact of substituting human decision-making with machine learning or organizational learning. In so doing, we relate how the key differences between human learning and machine learning may translate into lower within-organizational diversity and richness of background knowledge in organizational routines.
b) Our second contribution is to lay out important contingencies that affect the trade-off between the benefits of machine learning and the benefits of greater routine diversity, and knowledge richness and amplify or mute the aforementioned risks of myopia. By doing so, we hope to enable scholars and practitioners to achieve a deeper understanding of the organization learning-related risks due to machine learning.
c) Our third contribution is to highlight an important link between machine learning and routines that has not been emphasized so far, that the statistical variations in historical data arise from within-organizational variations in routines, and, hence, that model selections made by machine learning can be interpreted as selecting among routine variants.
Implications for Practice : Our article highlights the increased importance of the human element in mitigating organizational learning-related risks of machine learning. It suggests the requirement for governance mechanisms that focus on recognizing where the need for routine diversity and knowledge richness in organizations may be high, while also building alternative inventories of such diversity and knowledge that may be lost when machine learning replaces human decision-making.
Full Citations : Balasubramanian, N., Ye Y, and Xu M. Substituting Human Decision-Making with Machine Learning: Implications for Organizational Learning. Forthcoming at the Academy of Management Review.
Abstract : The richness of organizational learning relies on the ability of humans to develop diverse patterns of action by actively engaging with their environments and applying substantive rationality. The substitution of human decision-making with machine learning has the potential to alter this richness of organizational learning. Though machine learning is significantly faster and seemingly unconstrained by human cognitive limitations and inflexibility, it is not true sentient learning and relies on formal statistical analysis for decision-making. We propose that the distinct differences between human learning and machine learning risk decreasing the within-organizational diversity in routines and the extent of causal, contextual and general knowledge associated with routines. We theorize that these changes may affect organizational learning by exacerbating the myopia of learning and highlight some important contingencies that may mute or amplify the risk of such myopia.
- Substituting Human Decision Making with Machine Learning: Implications for Organizational Learning - December 7, 2020
- Human capital, parent size and the destination industry of spinouts - October 16, 2019
- Toward a dynamic notion of value creation and appropriation in firms: The concept and measurement of economic gain - September 14, 2016