Artificial Intelligence (AI) promises major performance gains in the public sector. However, as AI is primarily procured from private suppliers, adequate governance is crucial, particularly regarding AI risk perceptions. This study links existing AI governance frameworks and make-or-buy considerations to empirically examine perceived AI benefits and risks and their effects on supplier performance.
A novel workshare governance construct is established, capturing the allocation of digital development responsibilities and is adapted to a public sector setting, connecting dynamic capabilities with individual AI-related factors like motivation and risk awareness. A partial least squares structural equation model (PLS-SEM) is applied to test hypotheses between governance, AI readiness and performance with survey data from 104 questionnaires.
The necessity of workshare governance and its role as a mitigator of negative effects is empirically validated. AI risk awareness significantly reduces supplier performance, while AI motivation shows no significant effect. The results indicate that motivation alone is insufficient; risk awareness and governance determine whether AI leads to measurable improvements in supplier performance.
The study challenges the assumption that AI adoption automatically yields transformation. It highlights workshare governance and risk awareness as potential explanatory mechanisms, suggesting the need to refine existing AI readiness and adoption models, particularly for regulated or defense-related procurement environments.
Policy makers and procurement managers should prioritize governance frameworks addressing individual barriers to AI use and rigorously assess the value of potential AI products and services.
This study provides empirical evidence for the high relevance of procurement for public sector AI services via supplier-related governance perspectives.
