AI is becoming part of how we think, reflect, and make sense of the world. This changes conversation, not because AI is human, but because it can now enter the spaces where humans explore meaning together. The challenge is not simply to use AI well, but to remain responsible for judgement, relationship, and action while doing so.
We have always known, at least intuitively, that conversation is more than the exchange of information. A conversation is a social act. It is shaped by context and history, by the relationships among the people in the room, by emotion and timing, by what is left unsaid as much as by what is spoken.
When people talk, they bring not just information, but experience, vulnerability, memory, social stakes, and the possibility of being genuinely changed by the encounter.
So what happens when one of the participants in that conversational space is an AI?
This is no longer a theoretical question. Millions of people are already in daily dialogue with AI systems. They use them to think out loud, test ideas, explore problems, challenge assumptions, draft responses, reflect on difficult issues, and prepare for conversations they are not yet ready to have with colleagues or friends.
Increasingly, AI is also beginning to appear in group settings, as a resource in workshops, a support for collaborative sense-making, and even as a participant, of a kind, in collective inquiry.
This does not mean AI is a conversational partner in the same sense that another human being is. But it does mean that the conversational field is changing. AI is no longer simply something outside the conversation. It can now enter into the process by which we frame questions, make meaning, and decide what deserves attention.
Conversation as a space for thinking
Conversation is one of the primary ways human beings think together. We do not simply arrive with fully formed thoughts and exchange them. Often, we discover what we think in the course of speaking, listening, questioning, disagreeing, hesitating, and trying again.
This is especially important in complex situations, where the issue is not merely to find the right information, but to make sense of what is happening. In such situations, conversation is not a decorative extra. It is part of the work.
AI can support this work. It can extend our reasoning, broaden our perspectives, help us examine assumptions, and offer alternative ways of framing a problem. But it can also narrow our thinking if we use it carelessly.
The same is true of human conversation. A poor conversation can confirm prejudice, close down inquiry, and reinforce existing assumptions. A good conversation can open up possibilities, reveal tensions, and help people see more clearly.
The quality of our thinking depends less on the tools we have and more on how we engage with them, and with each other.
Too often, we treat both AI and people as sources of answers. We ask a question, receive a response, and move on. With AI, this can become a rapid prompt-and-response cycle with little reflection. With people, it can become polite agreement, quick debate, or the defence of fixed positions.
In both cases, the conversation closes too quickly. It becomes a shallow loop of confirmation rather than a genuine space for inquiry.
If conversations with AI are to add real value, we need to move from extracting answers to exploring meaning.
What AI brings, and what it lacks
The asymmetry matters.
A human conversation partner brings consciousness, lived experience, emotional resonance, embodiment, memory, and social responsibility. They have skin in the game. They can be surprised, hurt, delighted, embarrassed, challenged, or changed. They carry uncertainty and vulnerability into the room along with everyone else.
An AI brings none of that, at least not in any meaningful human sense. It does not have lived experience. It does not feel responsibility. It does not care about the consequences of what it says. It is not accountable to the group, the organization, or the community affected by the conversation.
But AI does bring something significant. It can synthesize large amounts of material, explore multiple perspectives, offer alternative framings, notice patterns, generate questions, and critique reasoning. It has no ego investment in being right, no status anxiety, and enormous patience.
It will not judge you for a half-formed idea. It will happily challenge your assumptions, offer counterarguments, suggest examples, and even critique its own critique.
These are not trivial gifts.
Perhaps the mistake is to ask whether AI is a conversational partner or merely a tool. It may be more useful to think of it as a new kind of cognitive participant within the conversational field. It can contribute to the thinking, but it cannot carry the human responsibilities of the conversation.
Meaning emerges in interaction
One of the most important things we understand from conversation, dialogue, and Knowledge Management is that meaning does not sit inside information waiting to be transferred from one mind to another. Meaning emerges through interaction.
Conversation is not simply a vehicle for sharing ideas. It is where ideas are shaped, tested, challenged, refined, and sometimes transformed. This is where intelligence often appears, not as something an individual possesses, but as something that emerges between people.
AI enters this space in a curious and potentially valuable way. It can help surface tacit assumptions, identify tensions, connect ideas across domains, and offer perspectives that might not otherwise arise in the conversation.
Used well, it can extend the conversational space and deepen inquiry. It can help people think more rigorously before bringing ideas into human dialogue, especially when they ask not merely for answers, but for critique, alternative interpretations, and possible blind spots.
But this also creates a danger. If AI provides a neat synthesis too quickly, people may mistake coherence for understanding. A fluent summary can make a messy conversation look more settled than it really is. It can smooth over disagreement, uncertainty, emotion, and minority perspectives.
In human affairs, meaning is often found in the very things that do not fit neatly into a summary.
The human work remains
AI can generate language, but it cannot take responsibility for meaning. It can suggest options, but it cannot decide what matters. It can summarize a conversation, but it cannot know what the conversation meant to the people in the room.
That human work remains with us.
We still have to ask: What is at stake here? Whose voice has not been heard? What assumptions are we making? What are we avoiding? What do we need to sit with a little longer? What would be wise, not merely efficient?
These are not simply analytical questions. They are relational, ethical, and practical questions. They require judgement, context, responsibility, and care.
This is why AI should not be allowed to become the hidden authority in the conversation. It can contribute. It can provoke. It can reflect. It can help us see what we may be missing. But it should not become the place where responsibility quietly disappears.
The more capable AI becomes, the more important human judgement becomes.
What changes in group dialogue
The implications for group conversation are especially interesting. When AI becomes part of a conversational space, whether as a synthesis tool, a reflective partner, or something more actively participatory, the dynamics of that space change.
Some of these changes may be positive. AI can help surface quieter perspectives, track themes over time, generate questions, bring in relevant material, and help a group notice patterns it may otherwise miss.
But these possibilities require us to facilitate differently.
In human dialogue, one of the facilitator’s primary tasks is to create conditions in which people can think and speak well together. This means attending to trust, pacing, power, safety, difference, and the emergent quality of collective thought.
All of that remains essential.
What changes is that we now also need to think carefully about the role of the AI in the room. What is it being invited to contribute? When is it useful? When should it remain silent? Who decides when it speaks? Can anyone ask for it to be paused or removed? Is it helping the group think, or is it quietly taking over the thinking?
These are not technical questions. They are conversational leadership questions.
There is also a real risk that genuine inquiry quietly collapses into information retrieval. The conversational question “What do we think?” can subtly become “What does the AI know?”
That is a profound change. It shifts the centre of gravity away from collective sense-making and toward the machine’s apparent authority.
The risk of premature closure
Human conversations are often valuable precisely because they are hesitant, incomplete, emotional, and uncertain. Meaning frequently emerges through pauses, disagreement, confusion, reflection, and the gradual evolution of understanding.
AI can unintentionally flatten some of this richness. Its fluency and confidence can encourage groups to move too quickly toward synthesis and closure. The conversation becomes cleaner, faster, and more efficient, but potentially less exploratory and less human.
This is not an argument against using AI. It is an argument for using it with greater awareness.
There are times when we want a summary. There are times when we want a challenge. There are times when we want alternative perspectives. But there are also times when we need to stay with the discomfort of not knowing, to listen more carefully, and to allow a conversation to unfold without rushing to resolution.
If AI is always available to synthesize, suggest, resolve ambiguity, and produce apparently coherent answers, we need to ask whether we are slowly losing some of our own capacity to sit with uncertainty and think things through together.
The point is not to reject AI, but to resist becoming dependent on it in ways that weaken our own conversational capabilities.
Working differently with AI
If we want conversations that genuinely augment intelligence, we need a different stance toward AI. Instead of treating it primarily as an answer machine, we can engage it as a reflective thinking partner.
The aim is not simply to get the right answer. It is to improve the quality of the thinking that leads to action.
That means asking AI to:
- challenge assumptions
- offer alternative interpretations
- identify what may be missing
- test where reasoning breaks down
- provide examples and counterexamples
- surface tensions or contradictions
- critique its own response
- help us formulate better questions
Used in this way, AI can help us see the limits of our own thinking. It can also prepare us for better human conversation, not by replacing that conversation, but by helping us enter it with sharper questions, richer perspectives, and a clearer sense of what still needs to be explored.
But we should not stop with the AI. Its contribution needs to be brought back into human dialogue, where it can be interpreted, challenged, adapted, rejected, or enriched in context.
Working differently with people
With human conversations, the conditions matter even more.
We need conversational spaces where different perspectives are invited, not merely tolerated. We need disagreement to be treated as useful rather than disruptive. We need people to be able to think out loud without always needing to be right.
Small shifts in questions can make a difference. Instead of asking “What do you think?”, we might ask “What might we be missing?” or “Who sees this differently?” or “What are we assuming too quickly?”
These kinds of questions open the conversation rather than close it down. They invite people to explore, rather than defend.
This matters because AI can sometimes make human contribution seem slower, messier, and less impressive. But the slowness and messiness of human conversation are not merely defects. They are often where judgement, trust, and meaning are formed.
We should not allow the smoothness of AI to make us impatient with the difficulty of thinking together.
Bringing AI into group dialogue
So, should we bring AI into group conversations? Probably yes, but thoughtfully.
Used carefully, AI may help groups reflect more deeply, widen perspectives, surface patterns, and improve collective sense-making. But the humans remain responsible for meaning-making, ethics, judgement, and relationships.
Perhaps this means we should:
- treat AI outputs as contributions, not conclusions
- make clear what role the AI is playing
- maintain space for uncertainty and disagreement
- resist moving too quickly to synthesis
- notice whose voices are amplified or diminished
- remain alert to the subtle transfer of authority to the machine
- keep asking what the human participants know, feel, notice, and question
- treat AI summaries as drafts for human correction, not as the official truth
This is where Conversational Leadership becomes even more important. If AI enters the room, someone still needs to hold the quality of the human conversation. Someone needs to notice when the group is becoming passive, when the AI is being treated as the authority, or when the pressure for a neat answer is closing down inquiry too soon.
AI may become part of the conversation, but it should not become the host of the conversation.
A practical rhythm
One useful rhythm may be to move iteratively between AI reflection and human conversation.
Explore a question with AI to stretch your thinking. Bring those perspectives into conversation with others. Test and adapt ideas in context. Then return to AI to reflect, challenge, and refine further.
This is not a rigid process. It is a way of keeping the thinking alive. It helps prevent the conversation from closing too quickly around easy conclusions, whether those conclusions come from people or from the machine.
The rhythm might look something like this:
- think with AI to broaden the frame
- talk with people to ground the issue in lived experience
- return to AI to test assumptions and explore alternatives
- bring the results back into human conversation for judgement and action
The crucial point is that AI reflection and human conversation should enrich each other. Neither should replace the other.
A new chapter in conversation
Conversation has always evolved as the conditions of human life have changed. The printing press changed what people needed to discuss. The telephone changed who could be in dialogue with whom. Digital networks changed the scale and speed of collective sense-making. AI is another step in that long story.
It does not end conversation, but it does change the conversational space in ways that are intellectually interesting and practically important.
For those of us who care about the quality of human dialogue, about the conditions in which people can think well together, share knowledge generously, and arrive at wiser collective understanding, this is not simply a technical issue. It is relational, social, ethical, and deeply human.
The opportunity is not that AI can think for us. It is that, used thoughtfully, it may help us think better together.
But the deeper challenge is this: how do we learn to think with AI while remaining fully responsible for our own judgement, relationships, and actions?
Closing Reflection
In-person, 7–11 September 2026
Warbrook House, Hampshire, UK
We are living and working in conditions of uncertainty, complexity, and rapid change. This week-long workshop offers a space to practise Conversational Leadership as a shared, lived experience.