A study examining what it takes for a company's AI project to succeed has been published in Journal of Computer Information Systems (JCIS, SCIE), an international journal in the field of information systems. The research was carried out jointly by KyeSook Kim and Professor Hyunchul Ahn of the Graduate School of Business IT at Kookmin University and researchers at the HJ Institute of Technology and Management (HJITM), with HJITM researchers serving as corresponding author.

The team surveyed participants in 249 completed AI projects at Korean companies, including project managers, AI developers, practitioners, and executives, and analyzed the responses using partial least squares structural equation modeling (PLS-SEM). They examined how top-management support, project champions, cross-departmental collaboration, an AI strategy aligned with business needs, and data availability and quality translate into project success.

The analysis showed that the factor with the greatest effect on AI project success was a clear AI strategy aligned with business needs (β=0.469), followed by data availability and quality (β=0.203) and top-management support (β=0.199). Project champions and cross-departmental collaboration had no direct effect on success, but contributed indirectly by strengthening the AI strategy and data quality, respectively. Cross-departmental collaboration in particular sharply raised data availability and quality (β=0.421).

Top-management support was the starting point of every path. Executive support boosted the presence of project champions (β=0.542), cross-departmental collaboration (β=0.497), and the AI strategy (β=0.281). The researchers interpret this to mean that strategy and business domain knowledge, more than data and infrastructure, determine whether AI projects succeed or fail.

The study is notable for empirically testing success factors using actual completed AI projects as the unit of analysis. It suggests that companies should not treat AI adoption as a purely technical task; instead, executives should take the lead in crafting an AI strategy tied to business goals and build a data foundation through cross-departmental collaboration. The findings may also inform government policies supporting corporate AI adoption.

HJITM continues to conduct joint research with scholars in Korea and abroad on organizational AI adoption and digital transformation.