Applied Research

Li, Hsing-Hsing - Lecturer, M.S. Technology Management

The integration of Artificial Intelligence into talent management use cases such as application tracking and video interviews introduces enterprise risk regarding regulatory compliance and ethical practice. Increasing global scrutiny, from the EEOC's investigations and FTC enforcement to emerging state/municipal algorithmic transparency laws, demands rigorous, auditable proof that hiring systems meet regulatory standards.

Research Objectives

1. Identify specific risk and compliance requirements specific to talent management systems and processes
2. Analyze industry recruiting workflows, user roles, and decision points to understand where human oversight is required, and where it is bypassed or impaired by the system design
3. Assess implementations, configuration, and model weaknesses and gaps, including model explainability and algorithmic bias
4. Benchmark leading HRIS and startup recruiting tech implementations, such as Workday, Rippling, Lever, Greenhouse, to compare instances of compliance and auditability

Hypothesis

Leveraging AI candidate scoring/ranking algorithms such as ATS and video interview systems create a governance gap, where user inputs and subjective preferences are often translated into automated decisions without adequate human review or auditable rationale. Although users often take the black-box nature of many AI candidate scoring/ranking algorithms at face value, algorithmic bias exists, exposing organizations to litigation risk.