How trust is built - and how it breaks
Trust is built in small steps. Every kept promise is one step. Every "I'll let you know by Friday" that actually gets said by Friday - that's one step. Every difficult piece of information delivered directly instead of wrapped in cotton wool - one step.
Small steps. A lot of time.
And here's the critical asymmetry: building requires dozens of steps. Destroying - just one. One unexpected move. One decision nobody understood and nobody explained. One moment when someone behaved differently than they always had.
It doesn't have to be bad behavior. Unpredictable is enough.
I've seen excellent leaders lose their team's trust not because they were dishonest - but because they changed their behavior under pressure. They started communicating decisions differently. Meetings stopped being places for open conversation and became theater. People stopped knowing what to expect.
And trust ran out.
Trust after layoffs
In organizations that went through reductions over the past two years, trust is a different resource than before.
Imagine your company in 2023 or 2024 laid off 20% of the workforce. Email from the CEO, meeting with HR, thank you for your contribution. Those who stayed watched. They observed how it was done. How beforehand it was all "we're one family" - and then came the lawyers' emails.
Those people are your team now.
They're not enemies. They're not bitter cynics. But they have new data about how the organization behaves under pressure. And that data affects how much they give, how directly they speak, how much they invest in professional relationships.
Rebuilding trust after reductions requires one thing above all: predictability. Not promises - because promises after 2023 sound hollow. Predictable behaviors, week by week, month by month, quarter by quarter.
When I work with teams after restructurings, I ask one question: "What would need to happen over six months for you to start trusting this organization again?" The answers are surprisingly simple. Not "give me a raise" or "give me security." Usually: "Tell me what you know. When you don't know - say you don't know. And do it regularly."
Predictability even in uncertainty.
Trust in the age of AI
This is a new context that organizations are only beginning to grapple with.
AI is entering team collaboration. It generates documents, summarizes meetings, analyzes data, supports decisions. And here an interesting problem appears: who do you trust when part of the work is done by an algorithm?
I observe two patterns.
First: "AI says X, so it must be true." Trust transferred to the tool without reflection. That's dangerous - not because AI lies. But because AI is unpredictable in a way that differs from humans. It doesn't lie deliberately - but it hallucinates. It doesn't judge context - it filters data. If you don't know what's going into the model, you don't know what to trust in the output. And as we've established, trust in an organization is built on predictability. A language model isn't predictable - at least not in the way we know from relationships with people.
Second pattern: "AI writes our reports, so I don't know who stands behind them." Trust between people dissolves because accountability becomes ambiguous. Who signs off on what the model generated? Who's responsible for an error in the analysis? If there's no clear answer - trust in both the content and the author drops simultaneously.
This isn't an argument against AI. It's an argument for making sure that introducing AI into team collaboration is deliberate and comes with clear accountability rules.
In organizations where I've seen AI adoption work well, one thing was consistent: people clearly stated what was their work and what was the tool's. "This is an AI-generated draft - I've verified the data and agree with the conclusions." Or: "The model proposed three options, I chose the second and here's why."
Transparency and accountability - that's the foundation of trust, whether you have five people and one model on your team, or fifty people and ten models.
Predictability as a practice
Back to Sebastian and his sentence.
He also told me: "Trust isn't something you declare. It's something you do for a year."
What does that look like in practice? A few behaviors I see in leaders their teams trust.
They keep small promises. Not just the big ones. If you said you'd come back with an answer on Tuesday - come back on Tuesday. If you can't - call on Monday and say you need more time. Small follow-throughs build the foundation on which big ones stand.
They communicate what they don't know. "I don't have a decision from the board yet. I'll let you know as soon as something becomes clear." Simpler than it sounds. And less common than you'd think.
They stay consistent under pressure. This is the hardest one. When things get difficult, when there's tension, when someone external is pushing - a leader who trusted their team has to keep trusting them. Or explain why they're changing their approach. Silent behavioral shifts under pressure destroy trust faster than open mistakes.
They admit mistakes. And don't make a production out of it. Plain and simple: "I was wrong in that assessment. I know more now. Here's what we're doing." No self-flagellation, no drama. Mistake + correction + move on.
The key that wears out
The title of this article is not a metaphor.
Trust, if you don't use it - if you don't subject it to tests, don't talk about it, don't consciously build it - genuinely wears out. It fades. Teams that once worked well, after a year of no investment in relationships, start becoming formal. Less honesty. More procedures. More CYA.*
The reverse is also true: trust that you actively build becomes an asset. A team that trusts you tells you things you wouldn't otherwise know. It responds faster in a crisis. Less energy goes to politics, more to actual work.
That's an investment with a real return.
CYA - cover your ass. Documenting everything in case things go sideways and you need to prove "that wasn't my fault." An organizational health thermometer: the more CYA, the less trust.
I'll leave you with the question Sebastian asked me at the end of our conversation - one I've since asked many others.
Does your team know what to expect from you - even in the hardest moments? And how do you know?