Acceptans för AI-verktyg i den offentliga upphandlingsprocessen

Sammanfattning

The purpose of this study was to explore the level of acceptance of artificial intelligence (AI) tools found in individuals working with public procurement in the public sector, what factors affect this acceptance, and how this acceptance can be affected by a brief intervention in the form of an introduction to an AI-tool. A qualitative approach was used in this study with semi structured interviews where users also received a brief introduction to Chat-GPT and were then allowed to test the tool themselves. Technology Acceptance Model 3 (TAM3) was used as a theoretical framework to identify which factors affect acceptance. The results indicate that acceptance for AI-tools in the public sector is generally high and is influenced by several factors, including experience, computer efficacy, and output quality. A brief introduction and practical experience with the AI-tool seemed to have led to an increase in acceptance, especially in individuals with low previous experience. Despite a high level of individual acceptance, organizational barriers, such as the absence of policy documents and clear guidelines, were identified as obstacles to actual use.