Experimental futuring: AI prototyping and pilot testing as anticipatory governance in education

The future impact of artificial intelligence (AI) has become a major policy concern in education globally. In this article, we identify how policy authorities are engaging in a form of anticipatory policy work that we term ‘experimental futuring’. Experimental futuring refers to the design of prototypes and pilot tests of AI applications that are intended to prefigure the possibilities of AI in the future. As an empirical case of experimental futuring in action, we examine a multi-year government programme of AI prototyping and pilot testing designed to address school improvement in England through in-depth documentary analysis. The analysis unfolds in two sections examining (1) proof of concept prototyping as pedagogic prefiguration, and (2) pilot testing as promissory evidence production. The result of this experimental futuring programme is a pathway to change that will sediment linear, promissory claims about AI’s future potential and ‘scalability’, and produce engineering blueprints for the future classroom, while foreclosing alternative possibilities for educational change. We argue it constitutes a distinctive mode of anticipatory policy action that is intended to prototype and test AI for scaling up both nationally and internationally, and prefigures a wider reorganisation of the English public sector for testing AI applications.