AutoML Fairness in HR Hiring: HCI Design Considerations

A recent thesis published on arXiv (2608.07477) investigates the concept of fairness within Automated Machine Learning (AutoML) tools utilized in human resource recruitment, incorporating insights from regulation, business strategy, and Human-Computer Interaction (HCI). The research posits that fairness is essential for usability, trust, legal adherence, and organizational acceptance, transcending mere ethical considerations. Although AutoML solutions facilitate model selection and implementation, they may inadvertently reinforce biased outcomes if trained on flawed historical hiring data. Many current platforms emphasize technical efficiency over fairness, which can hinder non-expert users from recognizing or addressing bias. The study explores four pivotal questions related to fairness mechanisms, interface clarity, human oversight, and design priorities, employing frameworks like the Technology Acceptance Model to assess user adoption. It underscores the necessity for HCI enhancements that foster fairness and improve user experience in AutoML systems for HR, asserting that fairness should be a fundamental aspect of the design process. This research is pertinent for developers, HR experts, and policymakers focused on ethical AI practices in hiring.