Lessons from the Public Sector Artificial Intelligence Adoption in Developing Countries: An AI Life Cycle Perspective

Artificial intelligence holds transformative potential for public services, yet its adoption in developing countries remains underexplored. This study examines AI adoption in Philippine public sector, addressing gaps in understanding challenges during later adoption stages in resource-constrained environments. Guided by the Technology-Organization-Environment (TOE) framework and Technology Affordances and Constraints Theory (TACT) and using a qualitative multiple case study based on semi-structured interviews with 10 key stakeholders from seven Philippine public sector AI projects, we analyze seven cases across the AI life cycle: design, development, and deployment. Findings reveal affordances and constraints influencing AI adoption in developing countries: strong leadership facilitates early design, but limited infrastructure, financial instability, talent attrition, and weak governance hinder development and deployment—challenges less prominent in developed contexts. This study contributes theoretically by integrating TOE and TACT with a life cycle perspective and empirically by uncovering AI adoption in developing countries, while offering actionable recommendations for practitioners.