Aligning Artificial Intelligence with Organizational Contexts (Volume 123) (English shop)

An Investigation of Natural Language Processing in Clinical Healthcare

Preview

Extract, PDF (310 KB)
Table of Contents, PDF (140 KB)

Artificial Intelligence (AI) is rapidly advancing in both performance and the range of tasks
it can perform, creating significant potential for innovative applications across various
contexts, including the healthcare sector. This progress provides solutions to pressing
issues, such as the shortage of healthcare professionals and the growing population of
older adults with increasing chronic diseases in many regions. However, effectively
aligning AI with organizational requirements presents significant sociotechnical
challenges. Misalignment can entail negative effects on organizations, including the
erosion of knowledge and employee dissatisfaction. However, as current studies are
fragmented across various disciplines and singular phenomena, their findings are often
not generalizable. This emphasizes the urgent need for a deeper understanding of the
sociotechnical components required for AI alignment.
This thesis addresses this problem by conceptualizing the sociotechnical components of
AI alignment as a multifaceted phenomenon, explored through a total of four studies. Two
studies explore the components of the phenomenon at a general level, providing an
overarching understanding. The remaining two studies specifically focus on the alignment
process, utilizing Natural Language Processing (NLP) as a case study within clinical
healthcare. The findings of this thesis offer meaningful implications for both research and
practice.
For researchers, the outlined framework on AI alignment enhances understanding of
contextualized AI and delineates the components that influence it from both
organizational and AI perspectives. Therefore, it enables researchers to position their
work within specific AI types and contexts, enriching the literature focusing on AI in
organizational contexts. Furthermore, the in-depth examination of NLP alignment in
clinical healthcare opens up novel research avenues that, as of yet, have been mostly
overlooked by the Information Systems (IS) community.
For practitioners, the framework serves as a valuable decision-making tool, guiding the
integration of AI into their (clinical) organizations and highlighting critical considerations
that must be addressed. Overall, this thesis contributes to a deeper understanding of AI
alignment and establishes a foundation for future research in this important and
dynamically evolving field.