Answers are grounded in this paper and cite their sources.
Public organisations now use artificial intelligence (AI) to answer citizen questions, support administrative decisions and automate routine work. Launching a system tells us little about whether service improved. This study asks under which institutional and governance conditions AI implementation produces measurable public-service outcomes and public value. It builds a framework that links AI readiness, implementation capacity, human-AI configuration, responsible governance, outcome-generation capacity, and measurement and evidence capacity. The design is a qualitative comparative multiple-case study using structured documentary analysis. Seven implemented cases make up the sample: GOV.UK Chat (United Kingdom), OuluBot (Finland), the 115 AI Service Chatbot (Germany), state-government chatbots (United States), several AI applications in Uganda, Service-Wise GPT (Nigeria) and SmartGov/NORA (Nigeria). A 31-variable coding framework scores readiness, implementation, governance, outcomes, public value and evidence quality. It keeps government-reported figures apart from independent evidence. Most of the governments in the sample show substantial implementation capability. Few show strong evidence of sustained public value. Four conditions recur across the cases: defined human-AI roles, bounded information environments, responsible governance, and the ability to measure and attribute outcomes. Deployment, implementation, operational improvement and public value need separate analysis. The study offers an evidence-oriented framework for tracing public-sector AI from institutional readiness to demonstrated public value. Keywords / tagsartificial intelligence; public administration; AI readiness; responsible AI; public value; digital government; human-AI complementarity; accountability; public-service outcomes
No takes yet. Share an insight, caveat, or question.
Oluwasegun Samuel Olawale (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: