In late 2022, ChatGPT rocked the world of education. ChatGPT is a freely available large language model AI system, which can generate writing across genres from a range of user created prompts. In recent months, social and mainstream media has been filled with stories about the ease of using ChatGPT to write convincing essays and other assessment tasks, raising concerns about academic integrity, including plagiarism. Universities and school systems responded, with some accepting its use and others banning it. These responses highlight a policy problem created by all forms of AI that are being introduced into education. That is, the current education policy environment lags behind the roll out of these technologies. Responses to ChatGPT also highlight the wide range of views on the use of AI, yet we have few methods for incorporating broad-based expertise and stakeholder input to create policies that support the productive use of AI while ameliorating potential harms. This paper contributes to current debates about the use AI in education by showing how collective policy making can serve as a method for creating more inclusive and participatory policy making (Emerson et al, 2012; Rickson & McKenzie, 2021) for AI in education. That is, we explore the potential role of participatory processes in (1) decision-making about the uses of AI in education (e.g., guidelines relating to the use of AI technologies to provide an education service) and (2) the uses of AI in education policy making and implementation (i.e., as part of education governance processes). The latter includes using AI to implement education policy (i.e., using automated systems to deliver high stakes tests) and to provide evidence to support policy making (e.g., new insights regarding links between inequality and student outcomes) (Gulson, Sellar & Webb, 2022). The paper outlines current approaches that can be used to enable collective policy making about AI through principled engagement, shared motivation, and capacity for joint action (Emerson et al, 2012). These approaches can include participatory procurement processes for education technology, policy prototyping, and education-specific algorithmic impact assessments (Gulson et al, 2022). These collective policy making methods aim to create ‘meaningful relationships between researchers and the different actors involved in the policy process’ (Rickson & McKenzie, 2021). The paper also explores the dynamics of new forms of collective policy making, which are emerging with the automation of ‘governmental decision-making’ (Paul, 2022) and the collaboration of machine and human policy actors.

Emerson, K., Nabatchi, T. & Balogh, S. (2011). An Integrative Framework for Collaborative Governance. Journal of Public Administration Research and Theory, 22(1), 1-29. Gulson, K. N., Sellar, S. & Webb, P.T. (2022). Algorithms of Education: How Datafication and Artificial Intelligence Shapes Policy. University of Minnesota Press. Paul, R. (2022), Can Critical Policy Studies Outsmart AI? Research Agenda on Artificial Intelligence Technologies and Public Policy. Critical Policy Studies, 16(4), 497-509 Rickinson, M., & McKenzie, M. (2021). The research-policy relationship in environmental and sustainability education. Environmental Education Research, 27(4), 465-479.