Learning about Markets from Models: Evidence from an AI Hiring Platform (arxiv.org)

A recent study explores the dynamics of an AI hiring platform that employs machine learning models to assess the retention likelihood of low-skill job applicants. The research reveals that while each firm benefits from its own tailored model, the models do not generalize effectively to other firms, highlighting a "model comparison externality." This differentiation allows companies to identify their unique competitive advantages and informs policymakers about how varied hiring strategies can be optimized. The significance of this study lies in its implications for both businesses and policymakers. Companies can leverage insights from the platform to refine their hiring processes based on distinct traits valued differently across firms, even in low-skill labor markets. The findings suggest that understanding these nuanced differences can guide future investment decisions and interventions. By pooling data across various contexts while maintaining unique models, the platform showcases how targeted hiring practices can enhance workforce retention, ultimately contributing to a more effective labor market.