2026-05-19T06:12:15Zhttps://keep.lib.asu.edu/oai/request

oai:keep.lib.asu.edu:node-2004662025-04-22T21:30:28Zoai_pmh:alloai_pmh:repo_items

200466 https://hdl.handle.net/2286/R.2.N.200466 http://rightsstatements.org/vocab/InC/1.0/ http://creativecommons.org/licenses/by-nc-sa/4.0 2025-05 49 pages Demas, Livia Balven, Rachel Pofahl, Geoffrey Barrett, The Honors College Thunderbird School of Global Management Department of Management and Entrepreneurship Department of Information Systems Over the past decade, Artificial Intelligence (AI) has permeated all aspects of society, including business operations, with undeniable benefits in terms of speed and consistency. In Human Resources (HR) departments, AI recruiting software is promoted as a revolutionary tool to streamline recruitment and improve candidate selection. However, growing evidence of biased algorithms has raised questions and concerns regarding AI models’ fairness. Through this thesis, I explored the hidden risk of bias perpetuation using both theoretical and practical lenses. The first section summarizes academic literature on the intersection between HR and AI and the origins of algorithmic bias. The second section presents a case study on IBM’s bias mitigation approach, developed through an in-person interview and follow-up communications with an IBM HR Tech consultant. The final section discusses similarities and gaps between the literature and IBM’s practices, concluding with applicable recommendations for companies. Research and companies both recognize bias as a core challenge in AI-assisted hiring. While scholars provide detailed and clear-cut mechanisms of bias perpetuation, companies such as IBM emphasize the need for preventive methods, noting that biases are difficult to identify and measure in real-life situations. This thesis highlights major gaps in the scope of AI implementation, insufficient training of both the technology and HR professionals, and the need for greater oversight and human involvement in AI development. Findings suggest that AI’s fairness is highly contingent on the developers’ and end users’ commitment to creating a bias-free hiring process. Human Resources AI Recruiting Software Bias Perpetuation Bias Mitigation Recruiting Tool Talent Acquisition The Hidden Risk of Bias in AI Recruiting Software: A Case Study of IBM’s Approach to Bias Mitigation