Generative AI is rapidly becoming part of managerial work, offering recommendations, analyses, and strategic advice across a wide range of organizational contexts. Although LLMs have the potential to improve productivity and support complex knowledge work, their performance remains inconsistent across tasks and domains. Consequently, determining when AI recommendations should be relied upon remains a major challenge for organizations.
Most existing research examines trust in AI as a relatively stable individual characteristic. However, managerial decisions differ substantially in terms of uncertainty, accountability, reversibility, and strategic importance. These contextual characteristics may fundamentally shape how decision makers calibrate their trust in AI-generated recommendations. Understanding these conditions is increasingly important as organizations integrate AI into high-stakes decision processes.

The thesis is supposed to answer the research question:
"How do decision characteristics influence managers' trust in and reliance on AI-generated recommendations during managerial decision making?"

If you are interested in this thesis, feel free to reach out to Tobias Kotzian (tobias.kotzian@unisg.ch).