Generative Artificial Intelligence in Human Resource Management: A Critical Systematic Review of Opportunities, Ethical Challenges, Governance, and the Future of Work

Generative artificial intelligence (GenAI), particularly large language models and multimodal foundation models, is reshaping human resource management (HRM) by extending automation from prediction and classification to content generation, conversational support, knowledge synthesis, personalization, and increasingly agentic workflow execution. This review critically synthesizes the emerging literature on GenAI enabled HRM, focusing on opportunities across the employee lifecycle, ethical and organizational risks, human–AI collaboration, workforce transformation, and governance. The review builds on the uploaded manuscripts four research questions concerning HR opportunities, ethical and legal risks, future of work implications, and responsible governance. Rather than treating GenAI as a stand-alone technology, the analysis conceptualizes value as an interaction among model capability, data quality, workflow design, human expertise, organizational culture, and governance.The review identifies a risk gradient: low risk, reversible activities such as drafting, summarization, learning content development, and routine employee support offer comparatively strong augmentation potential, whereas recruitment ranking, performance evaluation, promotion, compensation, disciplinary action, and termination require substantially stronger controls. A six-stage Responsible GenAI–HRM framework is refined around purpose and impact assessment, data stewardship, fairness by design, accountable human oversight, continuous monitoring, and workforce empowerment. The review further proposes research priorities involving measurable decision quality, employee trust, AI literacy, bounded autonomy, job redesign, cross cultural fairness, and sustainable adoption. The central conclusion is that GenAI should augment HR capability without transferring accountability for consequential employment decisions from people to machines.