Steven Pentland, an assistant professor in the Department of Information Technology and Supply Chain Management, was accepted in the Journal of Management Information Systems with his paper “Design Principles for Signal Detection in Modern Job Application Systems: Identifying Fabricated Qualifications”.
Pentland’s paper presents and evaluates a system for automated job candidate assessment during interviews. The system uses a highly structured interview paradigm and artificial intelligence (AI) to extract verbal and nonverbal behaviors from candidates. In the paper, the system is used to identify fabricated qualifications (deception), but the underlying design is intended to automatically assess a variety of behavioral metrics including personality and sociability.
Companies are already using artificial intelligence to select candidates, but academic understanding of these systems is lacking. His paper is a step towards conceptualizing the use of AI in the context of job candidate selection and assessing the strengths and weaknesses of this type of selection method.