Integration of Generative AI in the Recruitment ProcessCost Efficiency vs. Algorithmic Bias Risks in Tech

The adoption of Generative Artificial Intelligence (GenAI) and Large Language Models (LLMs) in Human Resource Management (HRM), particularly within recruitment and talent acquisition, has accelerated rapidly across global and national technology companies. This study critically analyzes the dialectic between the cost efficiency and process efficiency offered by GenAI versus the risks of algorithmic bias, transparency hurdles, data privacy vulnerabilities, and fairness in candidate evaluation. Utilizing a Systematic Literature Review (SLR) methodology guided by PRISMA standards across 23 peer-reviewed academic studies and reputable reports published between 2022 and 2026, this paper reveals that GenAI integration such as LLM-based semantic resume parsing, automated job description generation, and conversational chatbots can reduce time-to-hire by 40–60% and significantly lower operational hiring costs. However, GenAI models remain vulnerable to reproducing historical bias, engaging in proxy discrimination, triggering candidate information hallucinations, and creating opaque decision-making processes (black-box problem). This article formulates an ethical governance framework grounded in Human-in-the-Loop (HITL) oversight, regular algorithmic bias audits, data anonymization (blind screening), and Explainable AI (XAI) to ensure that operational speed does not compromise workforce diversity, equal opportunity, and workplace equity.