The meeting of AI and public-policy analysis is one of the most important new frontiers in democratic governance. For the first time, a small independent team — with no state funding and no apparatus of several hundred staff — can produce the kind of comprehensive, rigorous public-finance analysis that, until now, only government bureaucracies or large international organisations could deliver. We show what this looks like in practice.
Our analysis of Hungary’s 2026 national budget is the proof: 42 chapters, thousands of line items, structured data outputs, a ten-year renewal programme, and a section-by-section comparison with the Tisza party programme — all produced by a multi-agent AI system, under a single analytical framework. We are building a model that any civil-society organisation can apply in any country, for any budget cycle, provided it is committed to transparency and evidence-based reform. The tools of serious public-finance analysis must not remain a privilege of power.