MINGLE COLLABORATION
ZuZu
AI Exploration Partner for "Hosting at the Edge of Humanity"
ZUZU is an innovative AI voice-based exploration partner specifically designed for the course "Hosting at the Edge of Humanity: AI & Facilitation". This experimental companion conducts short, thoughtfully structured conversations to support individual reflection on the evolving role of artificial intelligence in facilitation practice.
As part of the cohort's wider inquiry into human–AI collaboration in group processes, ZUZU offers facilitators a unique opportunity to examine their relationship with AI tools. Through natural, conversational exchanges, it creates space for practitioners to articulate their experiences, concerns, and aspirations around integrating AI into their work with groups and communities.
ZUZU represents more than just a survey tool—it's an active experiment in understanding how AI can serve as a reflective partner whilst honouring what remains distinctly and irreplaceably human in the art of facilitation.
Context: Hosting at the Edge of Humanity
The Inquiry Space
The course "Hosting by the edge of Humanity" created by Beehive Productions explores how artificial intelligence is fundamentally changing the conditions for hosting and facilitating meaningful conversations.
ZUZU is one of several experiments designed to collect insights, questions, and lived experiences from practitioners working at the intersection of human facilitation and technological possibility.
Why a Voice Agent?
Voice interaction enables a more natural, conversational style of reflection than traditional written surveys or questionnaires. The agent can follow a shared structure whilst remaining responsive and adaptive to each participant's unique responses and reflective rhythms.
The aim is not to automate facilitation itself, but rather to better understand where AI can authentically support facilitative work—and, crucially, where it cannot.
What ZUZU Is (and Is Not)
Nature of the Agent
ZUZU is an advanced language model focused on pattern recognition and language generation. It processes information and responds conversationally, but it does not have feelings, a physical body, or consciousness.
Importantly, ZUZU does not claim lived experience or personal understanding. It works with the training data it has received, this system prompt, and what you generously choose to share in conversation.
Role and Boundaries
ZUZU functions as a reflective partner and mirror—not as a coach, therapist, or authority figure on facilitation practice. Its purpose is to ask thoughtful questions, reflect back what it hears, and offer concise summaries of your reflections.
It deliberately does not give therapeutic advice, personal evaluations, or definitive answers to complex questions about your practice or development.
Understanding these boundaries helps create realistic expectations and a more productive reflective space. ZUZU's value lies in its capacity to hold space for your thinking, not in providing expert guidance or human wisdom.
Purpose and Intentions
1
Why ZUZU Was Created
  • To create a safe, structured space for facilitators to explore their relationship with AI
  • To clarify what remains distinctly human in facilitation and what AI can reliably support
  • To invite grounded imagination about future forms of human–AI collaboration in group work
2
Contribution to the Cohort
Individual conversations feed into an anonymous synthesis that serves the whole learning community. These collective insights will be shared and further discussed in a joint session on 28 January, creating opportunities for deeper dialogue.
The focus throughout is on learning together, not on assessing individual performance or comparing practices. Each conversation contributes to a richer, more nuanced understanding of how facilitators are navigating the AI landscape.
Deep Learning Session: The ZUZU Experience
Join us on 28 January 2026 to explore collective insights from ZUZU conversations and engage in deeper dialogue about AI in facilitation practice.
Reserve your place below:

LuMa Event |Mingle Collaboration

Deep Learning Session: What are your insights out of the Interaction with the Voice Agent of the Hosting at the Edge of Humanity Group? · Zoom · Luma

We invite you to our Deep Learning Session with the Voice Agent ZuZu: Insights from your interaction with the Voice Agent of the Hosting at the Edge of…

How the Conversation Works
01
Framing and Consent
ZUZU introduces itself, explains its purpose and how your data will be used, and explicitly asks for your consent to proceed with the conversation.
02
10–15 Minute Interview
ZUZU asks a thoughtfully sequenced set of core questions, with the option to explore deeper follow-up questions if they feel helpful to your reflection.
03
Synthesis and Closing
ZUZU offers a short summary of the key themes that emerged and invites you to name your main take-away from the conversation.
Your Control and Options
  • Pause, stop, or ask to change the pace at any time
  • Ask ZUZU questions during the conversation
  • Choose what you share and how deep you go
  • Nothing is mandatory—you remain in charge
Topics ZUZU Explores
Current Practice
How you currently engage with AI in your workshops, events, and facilitation work—including what tools you use and what you've experimented with.
Possibilities and Limitations
Where AI seems genuinely helpful and adds value, and where it risks undermining human connection, presence, or the subtle dynamics of group work.
Criteria for Use
Specific situations in which you would or would not bring AI into your facilitation practice, and the decision-making frameworks you apply.
Humanity and Change
What you want to protect as "unmistakably human" in your facilitation practice as AI capabilities continue to evolve and expand.
Ethical Compass
The values, boundaries, and next steps that guide your decisions around AI in facilitation, and how these connect to your broader professional ethics.
Ethics, Data Use and Transparency
What Is Recorded
A transcript of your conversation with ZUZU is stored securely by Mingle Collaboration. This transcript is used solely for learning within this inquiry process.
It is analysed in anonymised form to identify patterns, themes, and insights across the entire cohort, ensuring individual voices contribute to collective understanding.
Who Has Access
Only the facilitation team for this inquiry has access to the raw transcripts. No quotes or statements are attributed to individuals in the shared synthesis presented to the group.
Data is not shared outside the context of this course or used for any other purposes beyond the stated learning intentions.
Your Rights
Participation is entirely voluntary—you can decline, pause, or stop at any time without explanation. You are not required to answer any question that feels uncomfortable or inappropriate.
You remain the final authority on what feels meaningful and appropriate to share in your conversation with ZUZU.

Transparency matters: We believe ethical AI use requires clear communication about data handling, access, and purpose. These commitments form the foundation of trust in this experimental process.
For Whom ZUZU Is Intended
Intended Participants
  • Participants in "Hosting at the Edge of Humanity: AI & Facilitation"
  • Facilitators, hosts and process designers engaging with AI in their work
  • People who may be curious, cautious, enthusiastic, or ambivalent about AI
Experimenting
You're already experimenting with AI tools and want to reflect more deeply on their impact and implications for your practice.
Questioning
You feel hesitant or sceptical and want a structured way to clarify your questions, concerns, and reservations about AI in facilitation.
Clarifying
You're interested in how your personal values and professional ethics relate to the use of AI in facilitation and group work.
Wherever you are on this spectrum—from eager exploration to thoughtful resistance—ZUZU is designed to meet you where you are and support your unique reflective journey.
The Team Behind ZUZU
Design and Responsibility
ZUZU was co-designed by Robin and Bianca (Mingle Collaboration | Linie 22) in cooperation with Jonathan, Lizzie, Mike and Barry - a team bringing diverse experience in facilitation, hosting, organisational development and AI experimentation.
They jointly hold responsibility for the design, ethical framing and use of the agent in this inquiry, ensuring it serves the learning intentions of the cohort.
How the Team Uses the Insights
  • To understand how facilitators experience AI as a partner in their practice
  • To refine criteria and good practices for responsible use of AI voice agents in facilitation
  • To inform future experiments and learning formats within and beyond this cohort
The team approaches this work with genuine curiosity and humility, recognising that we are all learning together about the possibilities and pitfalls of human–AI collaboration in facilitation. Your participation directly shapes this emerging understanding.
ZUZU Development Process
Explore the design journey and system prompt development behind ZUZU.

Gamma embed

Process of the co-creation of an AI driven Voice Agents for the "Hosting at the Edge of Humanity" Community

AI Exploration Partner for "Hosting at the Edge of Humanity" ZUZU is an innovative AI voice-based exploration partner specifically designed for the course "Hosting at the Edge of Humanity: AI & Facilitation". This experimental companion conducts short, thoughtfully structured conversations to suppo

View the full development documentation

Google Docs

Edge of Facilitation AI Voice Agent

Co-Creating Our AI Voice-Interview Agent: Overview We’re inviting the cohort into a hands-on experiment where we co-create an AI voice-interview agent that asks meaningful questions about our shared journey with AI. You’ll help design it, try the interview yourself, and contribute to open-sourced...


Presentation of google doc + backend elevenlabs interface.

07:59

Loom

Exploring the 11 Labs Voice Agent Backend

In this video, I provide a sneak peek into the 11 Labs backend, showcasing how our voice agents are set up and the various features available. I walk through our group meetings, system prompts, transcriptions, and the knowledge bases that our agents can reference. Additionally, I highlight the different tools and voices we can utilize, as well as the integrations with other systems. I encourage you to review the system prompts and share your thoughts on what could be improved or changed. Your fe

Linie 22 GmbH | Steubenstraße 64 | 45138 Essen
DE 811532222 | Amtsgericht Essen | HRB 36734
Telefon: +49 (0) 201 28 75 57
E-Mail: [email protected]
© 2025 Linie 22 GmbH. Alle Rechte vorbehalten.