Why Social AI?
AI is everywhere. You've probably already worked with it. Asked a question, floated an idea, turned a heap of thoughts into something you could actually use. A report, an analysis, a structure you couldn't see yourself.
And the questions that many people are grappling with: what does this mean for us? Will AI take over our work? Will the human element disappear?
Those questions keep me up at night too. And honestly, I don't think I have the answer.
But meanwhile, something else is happening.
What I see in sessions
What I notice in workshops, in long-running processes, is something different from the stories about AI taking everything over. I see AI that doesn't replace what's human, but amplifies it.
What if you applied that same artificial intelligence you use to answer questions to collaboration? Not AI drawing from its own sources, but AI working with what people said. With the words someone chose, because those words say something about what that person finds important.
That's where I want to go with this field guide.
From practice
A session with people from mental healthcare (GGZ, the Dutch mental health system). They're talking about transformation, about how care could be different. The facilitator is leading the conversation. A person with lived experience openly shares what he went through.
The conversation goes deep. About the tension between system requirements and human scale. Someone describes a woman with depression who actually just wanted to exercise. She couldn't afford ten euros a week for the gym, so she was referred to a psychiatrist. "The time I spent on that," the professional sighs, "she could have gone to the gym for six months with that."
The frustration is palpable. The system getting stuck. The rules that don't fit. The desire to change without budget cuts setting the tone.
But after forty-five minutes the conversation starts going in circles. The same question keeps coming back: how do we break through this?
"Show it, show it," someone says.
But how?
A silence. "I don't have it yet," one of the participants admits.
Then Jeroen, the facilitator, decides to try something. "Shall we see what the AI echo gives us? Maybe that'll help."
Behind that "echo button" is a prompt I had prepared beforehand. The AI reads the conversation and asks a question:
"Given the challenges you're describing, it seems crucial to start with small achievable steps that have direct impact in the neighborhood. Can we think of an example of such a concrete action that we could start tomorrow, without getting stuck in system requirements?"
From where I was sitting, I saw the group dynamic shift.
"This is quite something."
"That's the well-known, how do we start with something small tomorrow?"
"Nicely done!"
Suddenly concrete ideas emerged. Someone shared a project she was working on. Another person named specific actions for tomorrow. The question was there at the moment Jeroen needed it.
But here's what's interesting: it's not that AI asked a clever question.
What's interesting is what happened next: people felt heard. The question was based on what they had said. Their words, their concerns, their search, reflected back in a question that helped them move forward.
Why this moved me
What I saw in that moment was something I had been looking for a long time.
People who don't feel heard disengage. Or they participate but feel no ownership. Or they keep fighting to be heard, which gets labeled as resistance or getting in the way. Things we see continuously, in organizations, in teams, in our society.
And here I saw the opposite. A group that was stuck, that through a simple question (based on their own words) could suddenly move forward again. Not because AI gave the answer. But because AI reflected back what they had already said, in a way that helped them see the next step.
I think there's something fundamental here.
What this makes possible
AI can recognize patterns that we sense intuitively but can't name. Make visible what people find important, as individuals and as a group. Bridge the gap between what different people think and something that works for everyone.
Not by deciding. Not by summarizing. But by reflecting back what was said, in words that people recognize as their own.
This is what I mean by AI that amplifies what makes us human. Our creativity. Our desire to tell stories. Our need to be seen, to be heard, to matter.
And when you stack that, across sessions, across projects, across communities, something bigger emerges.
Imagine: local knowledge becoming scalable. An intervention that worked in a neighborhood. AI helps discover why it worked, how action led to results, how ownership grew over time. And then sharing that lesson, so another neighborhood, another municipality, another country can learn from it.
Collaboration at scale. Discovering more deeply what we need as people, in a street, in a municipality, in the world. What an individual needs. What we need together to be healthy and happy.
That's where I want to go.
That's what this field guide is about.
The first steps
It starts with capturing what people say. A transcript. That's the foundation, because then AI can work with it; not with its own knowledge but with the knowledge of the people themselves as raw material.
And then deepening. Seeing patterns. Confirming intuitions. Bridging the gap.
And then scale. From a session to processes over time. From a group to communities. From local wisdom to shared patterns.
That echo button from the story: why did it actually work? That's the question where the next part begins.