#22 Curiosity (Part 2 of 2) — The Price List

Somewhere around 1992 I got a Nintendo, and for the rest of the decade most of my free hours went into a screen: Super Mario on the NES and the Game Boy, later Half-Life, Counter-Strike and StarCraft on a PC. My mother and stepfather had a simple word for it: waste. A waste of time, and of a good brain.

Whether they were right is not the point. What matters is what the word "waste" did: it stayed with me long after the games. It taught me that being curious has a price, and that the people around you set it. Nobody banned the games; I just learned to keep quiet about them.

(This is the second of two editions on curiosity, the second of Microsoft's 5Cs covered here.)

8 min read. If you take one thing from this: interest is only half of curiosity. The other half is what it costs to be seen not knowing, and that price is set by the people around you. Ahead: the two halves, and why one without the other goes nowhere; who sets the price at work; what AI does to the wall, in both directions; which method fits which blocker.

Two halves

Start with a difference the word "curiosity" hides. Being interested in something and doing something about it are different things, and what separates them is rarely the interest.

Last week's edition covered the first: wanting the new, why it needs an information gap, and why the first weeks can feel like work. None of that explains why someone can be interested for years and never act.

Todd Kashdan's group split curiosity into five dimensions in 2018; two matter here. Joyous exploration is the interest itself: wanting to find out. Stress tolerance is something else: "the willingness to embrace the doubt, confusion, anxiety, and other forms of distress that arise from exploring". Put simply: can you cope with not knowing, and with being seen not knowing? In a representative US sample of 3,000, the group Kashdan calls Avoiders, about one in five, scored lowest on it.

So curiosity has two halves: interest, and the willingness to be uncomfortable in front of others. Microsoft's definition from the 5Cs, "what if we tried something completely different?", is written entirely about the first half. It says nothing about the meeting where you'd have to ask.

My own theory was that fear pushes curiosity aside when a person isn't strong enough to ignore what others think. Kashdan's model makes it clearer: the two are separate. Think of the colleague who has read every article about AI agents, can explain them at lunch, and has never opened one in a meeting. That is interest without stress tolerance.

Who sets the price

What my curiosity cost me, none of it dramatic:

In the nineties, being into computers was a niche, and even in 2005, when I started studying business computing at IBM, it was still seen as nerdy. When I became interested in coaching, early and seriously, the business world I worked in didn't treat it as a serious subject; support was thin. Other belief systems and practices were easier for people around me to laugh at than to ask about. And the games were waste.

Nobody in that list was hostile; the price is rarely a ban. It is a raised eyebrow, a joke, a "still on that?", repeated until you stop talking about it. Looking back: curiosity needs room, a place where the wrong interest costs nothing.

At work, the eyebrow has a name. Picture a team meeting three weeks into a new AI tool. Someone asks: "Sorry, what does this thing do with our data?" A pause. The next four seconds matter more than the rollout plan. If a colleague says "good question, I don't know either", the price of not knowing just fell for everyone there. If the team lead sighs and says "we covered this", it just went up, and the next basic question will probably not be asked. Amy Edmondson named this in 1999: psychological safety, "a shared belief that the team is safe for interpersonal risk-taking". A basic question about a tool the rest of the room seems to know is that kind of risk.

So the second half is a leadership question, not a personality trait.

A quick aside for HR and people-development leaders. If you're reading this and thinking that your culture, or your leadership, could use more curiosity, talk to me and we'll get you on the way. The second half is where that work lives: over months rather than days, as coaching, workshops and longer programmes, or a culture offsite as a starting point. If there's development budget left this year, this is the moment to decide whether it goes on another event your teams will have forgotten by February, or on getting one team moving. If you're planning next year, same conversation, only earlier and cheaper. I take on a few organisations at a time. Twenty minutes, and you'll leave with a clear next step whether or not you work with me: book a call.

Two sentences that stop people

The eyebrow isn't the only thing that stops people. Two sentences, I suspect, do much of the rest.

"I can't take on one more thing." True about time, attention and energy, not about the brain. It is not a hard drive that fills up; I know of no evidence for a storage limit.

"I'd only ever be a beginner at it." This one has a career attached: after years as the expert in the room, being a beginner feels like a step down. Loewenstein's information gap points the other way. You need to know a little before you can want to know more, and being a beginner is how the little gets there.

Capacity and status: both get weaker once you look at the research. The eyebrow doesn't. That one needs a room.

What AI does to the wall

Last week's wall was the grey phase at the start of something new, before you know enough to feel a gap. AI changes it in two ways.

The first is the one I'd argue for. It's eleven at night, the meeting is tomorrow, and someone types "what is an AI agent, explain it as if I'm new" into a chatbot. Nobody sees it, nobody sighs, the answer comes without an eyebrow. For someone whose interest has paid the social price for years, the first ten stupid questions are suddenly free. The price in the room is still set by the manager, and Microsoft's 2026 Work Trend Index, a survey of 20,000 knowledge workers in ten countries, suggests the two go together. Among the employees it calls "frontier", those most advanced with AI, 84 percent say their manager makes space to experiment, against 61 percent of the rest. A vendor survey, so a pattern, not proof.

The second has a name, given by BCG in June: distributed de-skilling, the loss of skills across a whole organisation. Their argument, from an executive survey (not peer-reviewed, so a warning, not a measurement): when AI replaces thinking instead of supporting it, judgement wastes away. Half the leaders surveyed already see it. Now picture that person six months later. Every question goes straight to the chatbot, the answer is the first and last step, and they have stopped forming questions of their own. Add Loewenstein's information gap and the mechanism is clear: if the wall disappears there is no gap to feel, and answers without effort create no urge to know more. A team that only ever gets the answer has had its curiosity answered for it, not trained.

So the design question for an AI rollout is not "how do we get people to use it" but "where in the wall do we want the AI?" My answer: at the start, for the first ten questions, where the social price is highest, then out of the way.

If you're the one meant to build it

For HR and people leaders: match the method to what is missing.

• Interest is missing. Last week's answer: a question from their own work, not a demo. Information gaps can be designed; feature lists cannot.
• Stress tolerance is missing. Protect the first attempts, in public. The most senior person in the room says "I don't know how this works, show me", out loud, and means it. That sentence does more for the price than a workshop can.
• Room is missing. Time, officially: an hour a week to try the tools, nothing to show for it. Reward the attempt, not only the outcome.

A training aimed at interest, given to a team whose problem is the eyebrow, tends to produce a polite room and no change.

The part I'll get wrong

Writing about human skills and AI every week is its own exercise in the second half: new territory, public mistakes. It took me about five attempts to share my thinking in public. Some of what I've written in these two editions I'll see differently in a year; better to say so now.

What changed for me since 1992 is not the interest but the price. Exploring used to feel like effort; now it feels like the point. That shift didn't come from more curiosity. It came, I think, from finding rooms where the eyebrow was gone.

Your turn

Your 1% this week. Pick the basic question about a tool or process you've never asked because everyone else seemed to know it.

• Ask it this week: a person if the room allows, a model if not. Notice whether you were avoiding the answer or the eyebrow.
• Then ask the next one in public.

If you lead a team, one more. In your next meeting, say "I don't know how this works, show me" about something real. Then watch the price for a month.

If you develop people, one more still. Put an hour a week on the calendar for people to try AI tools, with nothing they have to deliver. In the first session, ask the most senior person present to try something in front of the group and get it wrong. One visible mistake from the top shows everyone what a mistake costs here: nothing.

Next week: compassion, the third of the 5Cs. If you have a story of your own about the price of curiosity, tell me in the comments. I read every reply.

One favour before you close the tab. If you know a leader whose team has stopped asking about AI, or an HR colleague wondering why a good training changed nothing, forward this. Word of mouth is how most people find this newsletter, and a recommendation from you carries further than anything I can write.

If you're in HR or L&D and want to bring these themes into your organisation, here's that link again: book a 20-minute call. You can also explore the Leadership Essentials Workshop Series for teams.