| Abstract [eng] |
Generative Artificial Intelligence (GenAI) is transforming workplaces by shifting how organizations approach data analysis, insight generation, and problem-solving. While extensively adopted in technical fields such as software engineering and data science, its integration into non-technical roles remains underexplored. This research specifically investigates both the perceptions of non-technical employees regarding GenAI and the measurable effects of its use on their problem-solving capabilities in data-related workplace activities. The object of this research is the application of Generative Artificial Intelligence in supporting problem-solving processes among non-technical employees engaged in data-centric tasks. The aim of the research is to reveal how employees in non-technical roles perceive and apply generative AI tools to enhance their problem-solving capabilities while taking into consideration the organizational and contextual factors that support or prevent their effective use. An explanatory sequential mixed-methods approach was implemented, comprising a quantitative survey of 32 respondents, behavioral experiments involving 8 participants, and semi-structured interviews with 5 managerial representatives at a German manufacturing company. This methodological design enabled triangulation of findings across perceptual, behavioral, and contextual dimensions, strengthening the validity of the results despite the exploratory case-study scope. The main findings reveal significant barriers to effective genAI integration, including fragmented data systems, reliance on IT support for data access, cognitive overload resulting from complex AI outputs, and insufficient training in critical evaluation skills. Although survey responses indicated increased confidence and perceived efficiency when using genAI, particularly in creative, unstructured tasks, behavioral experiments demonstrated no measurable improvement in independent data reasoning or adaptive decision-making. These outcomes were assessed through task-based problem-solving exercises comparing performance with and without genAI assistance, revealing a clear perception-performance gap and highlighting the risk of over-reliance on AI-generated outputs. The study contributes to theoretical understanding by refining the Technology Acceptance Model (TAM), Diffusion of Innovations (DOI) theory. Contrary to TAM expectations, perceived ease of use was not a significant factor influencing adoption; instead, compatibility with existing workflows and trust in GenAI outputs emerged as critical determinants. This deviation is likely due to participants’ relatively high levels of digital familiarity, where functional value and contextual relevance outweighed usability concerns. Additionally, applying a Systems Thinking perspective demonstrated that task complexity moderates the perceived benefits of GenAI, with greater value observed in exploratory problem-solving rather than structured analytical activities. Based on these findings, several practical recommendations are proposed. Organizations should integrate genAI tools directly into established workflows to improve contextual relevance and functionality. Leadership must actively demonstrate responsible genAI use by participating in data driven decision-making processes and communicating clear ethical guidelines. Training programs should extend beyond technical onboarding to include structured critical thinking development, enabling employees to assess AI outputs rigorously. Furthermore, increasing genAI access to relevant organizational data is essential for improving the contextual accuracy of outputs, thereby strengthening user trust and supporting effective application in problem-solving contexts. In conclusion, this research advances the understanding of genAI adoption within non-technical roles, demonstrating that while GenAI tools enhance perceived problem-solving confidence, they do not inherently develop independent reasoning capabilities. Effective capability development requires not only thoughtful technological integration, but also targeted interventions aimed at strengthening critical thinking skills and improving access to high-quality organizational data. Given the exploratory case-study design and limited sample size, these findings should be interpreted within the context of the studied organization, with caution in generalizing to broader industry settings. |
