Algorithmic bias in AI-powered recruitment: Examining fairness, trust, and candidate experience in automated hiring systems

Recruitment has been one of the most automated tasks in modern human resource management. Most candidates and most jobs are now separated by resume screening tools, video interview analysers, gamified assessments and predictive scoring engines, often with neither party fully understanding how those systems operate. In this paper we explore perceptions and experiences of algorithmic bias in AI-powered recruitment through a survey of 224 job seekers and working professionals across a range of industries, and a focused literature review. The paper raises three related questions. How visible is algorithmic bias to the recipients of automated hiring decisions? What impact does perceived bias have on trust in employers and the recruitment process? And what conditions in the candidates’ eyes would make these systems feel more fair? The results reveal that concern about bias is widespread, but not universal. Respondents are especially uncomfortable with systems that employ facial analysis, voice analysis and historical hiring data, and are more comfortable with tools that perform narrow, well-defined tasks like scheduling or skills testing. However, when candidates believe that automated systems are being used without transparency, trust in employers falls dramatically, but bounces back if those same systems are combined with human review, clear explanations and a meaningful right of appeal. The paper argues that fairness in algorithmic recruitment is not a property that can be installed once and certified for all times. It has to be designed in, monitored continuously, and be supported by governance practices that take both technical and human dimensions seriously. The final chapters provide practical advice for HR leaders, vendors, regulators and educators in this fast-moving field.