To address the research questions, qualitative interviews were conducted with HR experts from Swissport and AI professionals. Furthermore, two GPT-based prototypes were developed and tested. The analysis demonstrates that GPT models can evaluate application documents in a structured and transparent manner and generate standardized scoring logic for subjective impressions observed during assessments.

The results confirm a strong potential for increasing efficiency, improving consistency, and enhancing fairness in the recruitment process. At the same time, key challenges were identified in the areas of data protection, ethical accountability, and user acceptance. The thesis recommends implementing GPT-supported tools gradually and transparently within a human-in-the-loop framework and integrating them into existing HR systems via appropriate interfaces. While the prototypes were developed based on a specific operational use case at Swissport, the underlying approach is transferable to other roles and industries. The work thus provides a practical contribution to the responsible digitalization of HR processes and offers strategic guidance for the future integration of generative AI into human resource management across different organizational contexts.