Barriers and Enablers for AI Adoption in Product Management: Implications for Project Initiation and Planning in IT Organizations

Sammanfattning

Artificial intelligence has become increasingly adopted across organizations, however its integration within product management remains underexplored and the implications of AI-enabled product decisions for project initiation and planning have not been examined. This study addresses these gaps by exploring how product managers perceive and experience the barriers and enablers influencing AI adoption, and how AI-enabled product decisions shape early-stage project activities in IT organizations. A small-scale qualitative survey design was employed with data collected through ten semi-structured interviews with product and project managers. The data was analyzed using a hybrid thematic analysis, combining deductive and inductive coding. The findings confirm that AI in product management is shaped by interconnected factors, consistent with prior AI adoption literature and established frameworks such as the Technology Acceptance Model (TAM) 2 and the Technology-Organization-Environment (TOE) Framework. This study also contributes with a number of original findings. Trust in AI is shown to be a dynamically constructed and task-specific practice rather than a static prerequisite for adoption. Additionally, two distinct motivational profiles, namely intrinsic and extrinsic, are identified where each one is associated with different adoption behaviour. A structural training gap is uncovered, showing that AI capability building within organizations is systematically directed towards engineering departments, while product and project roles are left without equivalent support. Furthermore, the normalization of ethical and privacy concerns through institutional delegation is shown to contradict prior literature that treats these as significant adoption barriers. In relation to project initiation and planning, AI-enabled rapid prototyping is identified as an emerging stakeholder communication medium that is replacing traditional documentation, and a quality-velocity tension is introduced as a new category of planning risk where AI-accelerated delivery outpaces the validation of customer value, a risk that existing project management frameworks do not anticipate. The findings suggest that companies cannot treat AI adoption in product management as a technological or organizational challenge. It requires deliberate attention to motivational dynamics, structural training equity, and the governance of AI-generated artifacts in early project phases. This study contributes an empirically grounded, cross-domain perspective that connects AI adoption in product management to early-stage project management, a connection that existing literature has not yet examined.