From Readiness to Regulation: Practice-Oriented Measures to Increase the Adoption of Generative AI in the Swiss Public Administration

This article outlines practitioner-identified and prioritized measures to support the adoption of Generative Artificial Intelligence (Gen AI) in the Swiss public administration. Despite growing interest, empirical guidance on how decentralized and compliance-oriented administrations can effectively increase the use of Gen AI remains limited, underscoring the need for practice-oriented research. Using a sequential mixed-method study design, a nationwide survey first assesses the use of Gen AI among Swiss public servants, followed by a Delphi study in which experts identified and prioritized measures to increase adoption. The results were mapped onto the Consolidated Framework for Implementation Research (CFIR) providing a theoretically grounded basis for categorizing the identified measures into key implementation domains. Survey findings show very low adoption rates, with experts highlighting “Ensuring access” and “Training” as top priorities to promote broader use. The CFIR analysis indicates that Gen AI implementation should focus on strengthening individual capacity and organizational readiness rather than relying on external mandates or purely technical adjustments. This has important implications for the sequencing of Gen AI implementation activities, especially in early-stage adoption phases.