The principles of deliberative democracy emphasise inclusivity, equality, and reasoned debate, with facilitation methods traditionally designed for in-person dialogue. However, the advent of digital deliberation platforms has raised critical questions about the extent to which AI-driven facilitation upholds or diverges from these foundational principles. This paper investigates the evolution of democratic deliberation through a comparative study of traditional in-person facilitation methods and emerging digital deliberation platforms, otherwise known as deliberative technologies.
The analysis is grounded in the author’s role as the independent evaluator of the 2024 Birmingham Museums Trust Citizens’ Jury (BMT CJ) in the UK. The BMT CJ represents a contemporary example of face-to-face deliberative processes, offering rich insights into how traditional facilitation methods foster connection, inclusivity and high-quality deliberation. These findings are contrasted with an evaluative review of two prominent digital deliberation platforms deployed internationally by Big Tech: the Stanford Deliberation Platform, employed by Meta for its Community Forums, and the Habermas Machine, developed by Google DeepMind. While both platforms aim to facilitate large-scale deliberation, their design choices raise critical questions around trust, inclusivity, user accessibility and the extent to which AI-driven facilitation upholds deliberative democratic values and quality of discourse.
Presented at the ECPR conference 2025, this paper introduces a novel framework grounded in the human-centred design principles summarised by the acronym SAAFE. It draws from findings of the research project funded by the Web Science Institute of the University of Southampton in the same year, in which the framework has been further developed to empower participants engaging with deliberative technologies.
This study employs a mixed-methods comparative methodological framework, combining qualitative case study analysis of the BMT CJ including surveys and interviews with participants as well as the facilitators, together with an evaluative review of the design, use, and outcomes of the two digital deliberation platforms.
Preliminary findings suggest that while digital platforms excel in scalability and speed of deliberation, they frequently struggle with issues of trust, equity, and quality of deliberation. The framework introduced in this paper provides a structured approach to addressing these challenges, emphasising the SAAFE design principles that prioritise both technological innovation and the human experience.
This paper contributes to the ongoing debate around the future of democratic engagement in a digital age by offering a systematic comparison of in-person and online deliberation, through the lens of facilitation methods and user experience. It highlights the need for collaboration between social scientists, technologists, and practitioners to develop hybrid models of democratic engagement that combine the strengths of in-person and digital deliberation, with reference to a set of guiding principles for its design and implementation.
The implications of this research are wide-ranging, extending to policymakers, politicians, platform developers and scholars of democracy across the world. As public deliberation and collective decision-making increasingly take place online, this paper calls for greater alignment between deliberative technologies and democratic principles. By adopting the SAAFE design principles, stakeholders can ensure that deliberative technologies empower participants while safeguarding the normative commitments of deliberative democracy in the 21st century.
