KI & Deliberation

Thu, 13 Nov 2025 00:00:00 +0000

Workshop: AI & Deliberation

How can generative AI improve public discourse and democratic participation?

The emergence of large language models (LLMs) opens up new opportunities to support and improve democratic processes and deliberation. But how can we harness this potential in concrete terms?

At the AI & Deliberation workshop, we want to approach this question in a practical way by presenting and discussing concrete tools and implementation approaches. Together, we want to learn about LLM-based applications that are useful for deliberative processes and discuss the opportunities they offer and their limitations.

The workshop promotes networking and knowledge transfer: by bringing in different perspectives, we want to learn from each other, exchange experiences and gain inspiration for new projects.

Target Audience

The workshop is aimed at researchers and practitioners:

Research & Science

Deliberation, discourse and argumentation research
Political science
AI community

Practice & Civil Society

Organisations that develop or want to develop LLM-based deliberation tools
Organisations that use and operate such tools in their civil society work

Tools

The following tools are presented and discussed:

reasonAide (EN): reasonAide bridges everyday conversations and online discourse. Instead of repetitive debates, reasonAide provides AI assistance to connect to a structured network of discussion points, understand perspectives, and participate constructively in public discourse.
IndI DiscussBot Demo (EN): An AI assistant for reformulating discussion posts from the IndI project. Formulations are analyzed in terms of politeness, reasoning, and empathy. If rudeness or a lack of empathy is detected, a suggestion for reformulation can be generated.
VISAR (EN): An AI writing assistant for formulating and revising discussion contributions. VISAR enables users to explore, test and validate their argumentative goals with the help of automatic draft prototypes.
PolicyCraft (EN): An LLM-based system that structures collaborative policy design through case-grounded deliberation. The tool supports users in collaboratively proposing, critiquing, and revising policies through discussion and voting on cases.
TITAN Chatbot (EN): The TITAN Chatbot engages users in interactive, Socratic-style dialogues designed to strengthen their ability to detect misinformation. It guides conversations through critical questioning, offers tailored explanations, and encourages reflection to deepen understanding.

For each tool there will be:

a short, roughly 25 min talk by a member of the developer/project team,
short, roughly 15 min commentary by an expert, and
a 35 min open discussion including all workshop attendees.

Schedule (all times CET/UTC+1)

Note: This schedule is preliminary and subject to change.

Thursday, 13th November 2025

Time	Session
2 PM	Opening & Welcome
2:15-3:30 PM	Tool session (EN): Policy Craft Tzu-Sheng Kuo (Talk, online) Julian Müller (Commentary)
3:30–4:00 PM	Coffee Break
4:00–5:15 PM	Tool session (EN): IndI DiscussBot Maike Behrendt (Talk) Franziska Maier (Commentary)
5:15-5:45 PM	Coffee Break
5:45-7:00 PM	Tool session (EN): VISAR Writing Assistant Zheng Zhang (Talk, online) Yannick von Ostheim (Commentary)

Friday, 14th November 2025

Time	Session
8:30 AM	Welcome Coffee
9:00-10:15 AM	Tool session (EN): reasonAide Michael Matheis (Talk) Eike Düvel (Commentary)
10:15–10:30 AM	Short Break
10:30-11:45 AM	Tool session (EN): TITAN Chatbot Dora Katsamori (Talk) Dennis Frieß (Commentary)
11:45-12:15 PM	Coffee Break
12:15-1:15 PM	Discussion (EN): Where do we go from here? Turning AI & deliberation into practice Jonas Fegert (Input & Moderation)
13:15-13:30	Closing Remarks

About the Discussion Session (12:15-1:15 PM): In this closing discussion, we will take a step back from the tools and reflect on what we’ve learned across disciplines from current prototypes of AI-assisted deliberation. Together, we’ll explore real-world needs, promising use cases, and how ideas can turn into future collaborations, projects, and possible funding initiatives.

Organisational Details

When: 13th November 2025, 2–7 PM & 14th November 2025, 8:30 AM – 1:30 PM
Where: Prinzessinnenstudio, Prinzessinnenstr. 16, 10969 Berlin
Language: German / English
Registration: kideku@itz.kit.edu Participation is free of charge. However, registration is required (limited number of participants for on-site participation). Hybrid/online participation is possible.

Why Participate?

Learn about the latest tools and prototypes
Network with experts from research and civil society
Discuss practical applications
Help shape the use of AI in a democratic context
Gain inspiration for new projects and collaborations

Opportunities of AI to strengthen our deliberative culture (KIdeKu)

Sat, 01 Jun 2024 00:00:00 +0000

In the KIdeKu project, we explored the question of how Large Language Models (LLMs) such as ChatGPT can be used to strengthen our deliberative culture. What opportunities do these technologies offer to improve our democratic practice? More specifically:

How can LLMs be used to improve the quality of public discourse?
Which current problems (disinformation, hate speech, …) can be addressed through the use of LLMs?
What requirements must be met for the successful use of LLMs?
Which concrete application scenarios are promising for a positive contribution of LLMs to deliberative culture?
How can we strengthen political participation through LLMs?
- For which target groups is the use of LLMs particularly promising?
- What particular challenges does the target group of young people pose?
…

Goals & procedure

👥 Development of deployment scenarios: In consultation with civil society actors, we designed relevant and novel deployment scenarios. We wanted to create a broad overview of how LLMs can be used in our democratic practice for the common good and of the goals that can be pursued with them. We considered different requirements by civil society organizations for the use and integration of LLM-based applications as well as the needs of the various target groups. Examples of such application scenarios include:
- Fact checkers, hate speech detectors, AI assistance in writing speeches, (partially) automated moderation in online debates, debate summarization, argumentation explanation, argument checking, …
🤖 Creation of technical foundations: For some of the conceptualized deployment scenarios, we developed technical foundations that can be used and further developed by the community. These foundations include:
- Requirements that can be used to systematically test the suitability of language models for the intended use. These requirements can be operationalized in the form of deliberative benchmarks (test data sets). Demo apps that illustrate the use of LLMs and can be the starting point for the development of ready-to-use apps.
📋 Action orientation: A final report summarizes the current state of knowledge on the use of LLMs in deliberative democracies to strengthen our deliberative culture, compiles project results and formulates recommendations and best practices for civil society and political actors.

Overall, in KIdeKu we developed concepts and studies to create a suitable operational framework for the development and use of AI for the common good in our democratic practice. We hope that the results and ideas will be taken up and further developed.