Events

Mr. Shaoqin Tang
Ph.D. candidate in Strategy and Entrepreneurship
Leeds School of Business
University of Colorado Boulder

While traditional literature emphasizes the importance of deploying high-capability workers to tasks, this paper argues that the rise of algorithmic technologies enables the configuration of human-algorithm capability bundles, altering how organizations deploy workers with varying capabilities. Using a formal model, I show that the effects of algorithms on worker deployment depend on the locus of human capability and the trajectory of algorithmic capability advancement. Consider tasks comprising two complementary components. Low-capability workers underperform relative to high-capability workers in both components, but the performance gap is larger in component 1 than in component 2. When algorithmic capability advances in component 1 (“algorithmic deepening”), organizations benefit from bundling algorithms with low-capability workers to perform tasks, reducing their need for high-capability workers. Conversely, when algorithmic capability extends into component 2 (“algorithmic broadening”), organizations substitute low-capability workers with algorithms, while maintaining demand for high-capability workers. I examine these predictions by tracking the deployment of reviewers on firm-hosted open-source software projects over a period when the algorithm used in code review first deepens and later broadens its capabilities. My study has important implications for capability bundling and organization design in an age where humans increasingly interact with algorithms in task performance.