Bortom traditionell teknikadoption : En fallstudie av organisatoriska, individuella och AI-specifikafaktorer vid implementeringen av generativ AI

Sammanfattning

Generative AI has rapidly become a central part of organizational digital transformation and created new opportunities for efficiency and knowledge-intensive work. Unlike previous technologies, generative AI can function as an active partner in work processes, raising questions regarding how established technology adoption models can be used to understand its implementation. The purpose of this study is to investigate whether the Technology-Organization- Environment (TOE) framework, the Unified Theory of Acceptance and Use of Technology (UTAUT), and the Threat of Artificial Intelligence (TAI) perspective require an integrated approach to explain the adoption of generative AI, or if the models possess sufficient explanatory power individually. Through a qualitative case study of the e-commerce company Lyko's pilot project with Claude Teams, the study examines potential alignments or perception gaps between strategic vision and operational reality. Furthermore, it evaluates whether generative AI introduces unique, AI-specific factors compared to previous adoption patterns for cloud services and RPA. The empirical data consisted of structured interviews with individuals at both strategic and operational levels within the organization, as well as a survey distributed to participants in the pilot project. The analysis was conducted through a deductive thematic analysis based on TOE, UTAUT, and TAI. The findings show that the adoption of generative AI is influenced by both organizational and individual factors, such as management support, technological readiness, perceived usefulness, and social influence. The study also identifies perception gaps between management’s strategic vision and employees’ operational experiences of the technology. Furthermore, the results indicate that certain AI-specific aspects, such as uncertainty regarding the technology’s impact on work processes, are not fully captured by traditional technology adoption models. The study concludes that an integrated understanding of TOE, UTAUT, and TAI provides a more comprehensive explanation of generative AI adoption than the frameworks do separately.