A lot is being said about AI, and of course, a lot of it is true, and a lot of it isn’t. I’m the first to admit that I simply love AI. I joke that I have a not-so-secret lover these days, and his name is Claude. But I also have to admit that my view on how to use AI has changed. The reason I fell deeply, madly in love with AI is that I moved beyond that first experimental stage and finally cracked how AI can actually help me get better. How to move from output to outcomes.
You know, when you first start using AI, you’re blown away by its ability to write stuff for you. But, like many of us, I very quickly got tired of the AI-generated texts popping up everywhere. And yes, you can give it tools, instructions and skills to make the writing better, but that isn’t what got me hooked on Claude.
What got me hooked was the shift in how I use AI: from ‘it does my writing’ to ‘it does my analysis, and I still have to bring my experience, judgement and critical voice to get to the level I’ve been yearning for for years’. That changed everything. All those plans I had. All those ideas. All those complicated analyses. AI has sped them up enormously, and in some cases made them possible in the first place. So no, it isn’t taking over my thinking. It is helping me connect the dots and untangle the knots. It has become the rocket engine underneath me. It has opened up possibilities I can now venture into because I’m no longer restricted by my ability to code, or by the speed and cost of analysis and research.
But I notice that the fear of using AI is getting the overhand of the fact to see and discover with an open-mind what AI could bring or add. Not just replace. And to experience this, people need to be able to play around with AI a bit, just like I did. Even if you are a novice or a non-expert. I noticed lately that very capable people went quiet when AI came up, each clearly worried they were the only one who had not figured it out yet. It made me think about an interesting design question: How do you design an organisation where people dare to use AI?
In searching for the answer, I kept returning to a single, telling pattern. The conventional response to slow AI adoption is to roll out more tools and more training, on the assumption that the gap is one of skill. But the real blocker is usually social. In most teams, everyone privately feels behind, assumes everyone else has it figured out, and so no one admits they are unsure, no one asks the basic question, and the whole group quietly avoids the thing. The fix is not another course. It is a context where not knowing is openly normal: a space, physical or virtual, an AI playground, where experimenting is the point, where the explicit social norm is that everyone is starting from zero. The behavioural science here is about a strange trap a group can fall into together.
When everyone privately doubts something but assumes everyone else is fine with it, the group as a whole acts on a belief that almost no individual holds.
Psychologists call this pluralistic ignorance, and it is exactly what freezes AI adoption: a room full of people each thinking they are the only one who is lost, all staying silent, all confirming each other's false impression that everyone else has it sorted. The silence is not a lack of ability.
It is everyone protecting themselves from a judgement no one is actually making.
This is Behavioural Design applied to a collective fear. You do not break the silence by telling people to be braver or sending them on a course. You change the context so that not knowing is openly the norm, and once it is safe to be a beginner, the questions, the experimenting and the learning come flooding out. So, how do you put this to work? Here are three practical moves:
Make not knowing explicitly normal. The block is that everyone thinks they are the only one who is unsure. Have the most senior person say out loud that they are figuring it out too, and you puncture the illusion that everyone else has it sorted.
Create a space where experimenting is the point. People will not risk fumbling with a new tool in front of an audience that seems to know more. Give them a low-stakes playground where messing about is expected, and the learning starts.
Set the starting line at zero, for everyone. When you introduce something new, say plainly that nobody is expected to be good at it yet. Naming the shared starting point removes the fear of being the one who is behind.
I have had so much fun experimenting with AI. But you know what I really like? Sometimes it is so confidently wrong. And I can say so right to its face, with no repercussions at all. Even better is when it comes back with: "You are right, Astrid." Oh, how I love the sound of that. Point being: your input, and your ability to judge what AI gives you, is exactly what makes it less scary. Combine that with a thinking partner that has no ego and no thin skin, and honestly, that is rather wonderful.
0,5 Minute on, Influence
Social loafing: why people pull less in a crowd
You have probably sat in a large group project where, somehow, less got done per person than a small team would have managed. There is a name for this: social loafing,
The tendency for individuals to put in less effort when their contribution is pooled into a group and cannot be seen.
Researchers tracing it back to Maximilien Ringelmann found people literally pulled less hard on a rope as the team grew, each quietly easing off. The reason is that when effort is invisible and responsibility is shared, the link between what you put in and what is noticed gets weaker, and effort drifts down with it.
The fix is to make individual contribution visible again.
People work harder when their part is identifiable, so the move is not to motivate the group but to design the work so each person's contribution can be seen: clear ownership of distinct pieces, smaller teams, roles that do not blur into an anonymous collective. When everyone is vaguely responsible, no one quite is, so give each person something that is unmistakably theirs.
See you next week!
Astrid Groenewegen
Co-Founder of SUE | Behavioural Design Academy, Author of 'The Art of Designing Behaviour' / 'De Kunst van Gedrag Ontwerpen' and ‘De Gelukscode’.
Source:
On pluralistic ignorance: Katz, D., and Allport, F. H. (1931). Student Attitudes. Craftsman Press. Short piece: Ringelmann, M. (1913), on social loafing (the Ringelmann effect).