The technology transfer office (TTO) emerged as an ad hoc institutional response to a specific American legal and economic moment: Niels Reimers's founding of Stanford University's Office of Technology Licensing in 1970, and the subsequent enactment of the Bayh-Dole Act in 1980, which granted universities title to inventions arising from federally funded research. Over five decades, this single national model diffused globally as the default architecture for university commercialisation, largely without adaptation to the radically different capital markets, institutional maturity, and political economies of the countries that later adopted it. This paper traces that diffusion; situates it within longstanding philosophical debates over the commercialisation of academic science, from Mertonian norms through academic capitalism, the Triple Helix, and contemporary epistemic-justice critiques; and argues that the TTO's persistence as a near-universal template has become a structural liability for catching-up innovation systems, particularly in Eastern Europe, where output-side innovation capacity now consistently outpaces input-side commercialisation infrastructure. Drawing on a twelve-scenario typology of alternative and hybrid commercialisation entities, together with an explicit multi-criteria method for scoring institutional feasibility under resource constraints, the paper argues that countries lagging in technology-transfer infrastructure are better served by choosing deliberately among a design space of models than by importing the Stanford template wholesale.