Generative artificial intelligence (AI) requires data as an input. The way data is maintained and managed within an organization is determined by Data Governance (DG). This master thesis addresses the question of which DG elements are necessary to ensure the sustainable, profitable, and long-term application of generative AI. The thesis aims to provide a Data Governance Framework (DGF) tailored to Swiss MSEs in the manufacturing sector.

To answer this question, the literature was analyzed to identify existing DGFs. Their applicability was evaluated, hurdles in the application were described, and missing elements were identified. Further, the current situation of Swiss MSEs in the manufacturing sector was methodically described through surveys and expert interviews. Insights from this research led to the development of eight Critical Success Factors (CSFs) outlining the aspects of Data Governance essential for the successful application of generative AI. Incorporating these CSFs and consulting experts in generative AI, IT security, Data Governance, and Legal Compliance, the thesis identified key Data Governance Elements and designed a DGF for the sustainable application of generative AI by Swiss MSEs. An expert reviewed this framework, and initial feedback was integrated into a revised version. This master thesis develops a DGF that enables Swiss MSEs in the manufacturing sector to apply generative AI applications sustainably.

The findings confirm that organizations can ensure the sustainable use of generative AI by leveraging specific elements of Data Governance.