PhD project Behind every AI system are human decisions: what data to use, who gets to decide, and which judgments are trusted. My doctoral project examines how organizations make those choices and turn them into working AI systems. Through two real cases in the public sector, it highlights how everyday organizational practices determine whether AI becomes a helpful tool or a risky black box.
My doctoral project examines how organizations create the knowledge foundations that AI relies on. Data is often portrayed as given or the ''new oil'', yet it must be interpreted and validated by people. Studying healthcare and road maintenance with municipalities, I explore how these public organizations without stable data definitions decide what counts as trustworthy data, manage uncertainty and balance expertise with technology. The results aim to improve data governance and support reliable AI developments.
This doctoral project examines how organizations create, negotiate and maintain the knowledge foundations that artificial intelligence (AI) relies on. Although data is often described as a natural resource ready for use, research in information systems shows that data must always be collected, interpreted and structured by people working within specific organizational and technological settings.
The project investigates two empirical cases where organizations aim to develop or apply AI systems but struggle with ambiguous definitions of what counts as correct data.
The findings are expected to contribute to information systems research on data governance, sociotechnical systems and AI development practices. More broadly, the research offers practical insights for organizations seeking to align AI solutions with real operational needs, improve data quality and develop trustworthy and sustainable AI systems.
