Institutional Design for AI-enabled Deliberation

Abstract

AI applications such as summarization and translation have the potential to enable mass deliberation. However, governments need institutional and practical guidance to implement AI features in deliberative processes that are still meaningful. The AI4Deliberation project aims to design both practical AI tools and comprehensive guidance to support governments in their eforts towards AI-enabled deliberation. In this workshop, we want to explore the requirements and conditions for the institutional context of AI-enabled deliberation.

Keywords

• Deliberative democracy
• artificial intelligence
• institutional design
• practical guidelines

1. Introduction

[ 3 ]. However, the implementation of a well-functioning summarization tool into a deliberative platform does not guarantee a meaningful, sustainable, and ethical deliberative process. On the other hand, the actual implementation of AI in deliberative practices has not yet been experimentally tested widely.

Authors stress the importance of the design of deliberative processes, and of a socio-technical perspective on AI-enabled deliberation [ 4, 5 ]. Following Koppenjan and Groenewegen [ 6 ], these two can be brought together by first establishing the structure of the deliberative process. Thereafter, the technological and institutional artefacts that can instigate and shape this process should be congruently designed. Institutional design comprises all social rules that structure the behavior of actors in the deliberative process [ 6 ] and has always been a critical aspect of deliberation [ 4 ].

2. AI4Deliberation project

The Horizon project AI4Deliberation aims to explore the technological and institutional aspects needed to realize the potential of AI-enabled deliberation [ 7 ]. The project aims to provide robust, ethical AI tools that support governments in achieving multimodal, gamified, mass deliberations, resulting in more active and inclusive citizenship and increased trust in the rule of law order by citizens. The AI tools

CEUR Workshop

ISSN1613-0073 will be accompanied by comprehensive guidance to assist government in institutionalizing, designing, and using AI tools in deliberative processes.

This guidance is delivered through a comprehensive framework that prescribes the institutional context that ensures efective, safe, and meaningful enactment of AI-enabled deliberation in the public domain. The framework will comprise practical guidelines, policy recommendations, road maps, investment decisions, and training material for capacity building. The framework will be developed through co-creation and deliberation. A first step is to define conditions and requirements for the institutional design of AI-enabled deliberation.

3. Workshop objective and program

This workshop aims to gather researchers to explore the implications of this new topic and to discuss the conditions, requirements, and challenges related to institutionalizing AI-enabled deliberation. We will start with introducing the AI4Deliberation project and the comprehensive framework, but will shortly turn to an interactive co-creation session. We invite researchers from a broad range of disciplines to share their perspectives on this novel and relevant development in democratic practices and digital government. We will share our insights on AI-enabled deliberation gained in our project so far. But, most of all, we hope to collectively further the knowledge on this pressing topic.

Acknowledgments

Funded by the European Union under grant agreement No. 101178806. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Research Executive agency. Neither the European Union nor the granting authority can be held responsible for them.

Declaration on Generative AI

The author(s) have not employed any Generative AI tools.

References

[1] A. Gutmann , D. F. Thompson , Why Deliberative Democracy?, Princeton University Press, Princeton, NJ, 2004 .

[2] N. Alnemr , Emancipation cannot be programmed: blind spots of algorithmic facilitation in online deliberation , Contemporary Politics 26 ( 2020 ) 531 - 552 . doi: 10 .1080/13569775. 2020 . 1791306 .

[3] R. Shortall , A. Itten , M. van der Meer, P. Murukannaiah, C. Jonker , Reason against the machine? future directions for mass online deliberation , Frontiers in Political Science 4 ( 2022 ). doi: 10 .3389/ fpos. 2022 . 946589 .

[4] D. Friess , C. Eilders , A systematic review of online deliberation research , Policy and Internet 7 ( 2015 ) 319 - 339 . doi: 10 .1002/poi3. 95 .

[5] J. A. Carstens , D. Friess , Ai within online discussions: Rational, civil , privileged?, Minds and Machines 34 ( 2024 ). doi:10.1007/s11023- 024- 09658- 0.

[6] J. Koppenjan , J. Groenewegen , Institutional design for complex technological systems , International Journal of Technology, Policy, and Management 5 ( 2005 ) 240 - 257 . doi: 10 .1504/IJTPM. 2005 . 008406 .

[7] E. Tambouris , Artificial intelligence for mass democratic deliberations: The ai4deliberation project objectives and methodology , in: Proceedings of the 13th. Computing Conference , 2025 . [accepted for publication].