AI Interviews and Strategic Applicants
讲座信息 | INFOMATION
讲者:郑权教授(中国科学技术大学)
时间:2024年12月6日(周五) 10时
地点: 同济大厦A座401教室
讲座摘要 | ABSTRACT
An increasing number of firms are adopting artificial intelligence (AI) to interview and assess job applicants. In this research, we develop a game-theoretic model to examine an AI provider's decisions on algorithm design and pricing for AI interview products. We also investigate when hiring firms should adopt AI interview products and how this adoption decision changes with the nature of their job. We consider jobs requiring hard skills (e.g., technical expertise) and those requiring soft skills (e.g., personality), with the latter being more prone to manipulation during interviews. AI's precisions in assessing either skill are endogenously determined, and pre-determined hiring rules are used to screen applicants. By contrast, HR interviewers can better assess personality because of human instinct but have limited ability to assess expertise, while considering applicants' incentive to manipulate personality when making hiring decisions. We present three main findings: First, even if improving precision rates is costless, the AI provider strategically reduces the precision of its AI interview product in assessing either personality or expertise to mitigate applicants' personality manipulation. Second, firms are more likely to adopt AI interviews for lopsided jobs that rely predominantly on either personality or expertise. However, for balanced jobs that rely moderately on both skills, firms prefer AI interviews when HR interviewers are not skilled at assessing expertise, or when HR interviewers are skilled at assessing expertise but personality manipulation is highly costly for applicants. Third, the AI provider can sell the AI interview product at higher prices for jobs that rely more on personality despite the fact that HR interviewers surpass AI in evaluating personality and accounting for manipulations.
讲者介绍 | GUEST BIO
郑权,佛罗里达大学博士,现任中国科学技术大学管理学院特任教授。主要研究领域为人工智能经济学,零售和平台运营管理,行为学定价,供应链管理,运营管理和市场营销/信息系统交叉学科研究等。相关研究成果发表在Marketing Science、Management Science、M&SOM、Production and Operations Management、MIS Quarterly 等知名期刊上。