Polyconomy seeks to use a fine-tuned Large Language Model (LLM) trained on classical-liberal and heterodox economic literature.

Its goal is to provide high-quality, well-sourced answers for academics, journalists, policymakers, and the public—free from the narrow assumptions of mainstream economic models.

The AI will draw on insights from traditions such as Austrian, Public Choice, Monetarism, New Institutional Economics and from other traditions in Classical Liberalism, building a diverse repository of knowledge. Emphasising intellectual pluralism and ideological transparency, the project supports research, teaching, and policy analysis.

Recently, Joshua Bowden joined the Vinson Centre to train the model on the relevant texts, working with the coders at the School of Computing to deliver the best possible result. Work continues apace with more updates on the project in the not too distant future.