AI Didn’t Break Management
AI didn’t make management harder. It made weak management visible.
This piece looks at how effort, busyness, and over-involvement stopped being enough — and why clarity, judgement, and restraint now matter more than ever.
Most organisations don’t suffer from a lack of decisions.
They suffer from decisions that never quite land.
They circle.
They stall.
They get revisited.
They resurface in new meetings wearing different clothes.
And everyone feels it.
Decision hoarding rarely looks dramatic.
It doesn’t show up as control or micromanagement.
It shows up as:
“Let’s just sense-check this first”
“I want to stay close to it”
“Run it by me before we commit”
“I’ll make the call once I’ve seen the next version”
Nothing is explicitly blocked.
Nothing is fully released either.
So work keeps moving — slowly, cautiously, and without ownership.
Decision hoarding isn’t usually about ego.
It’s about risk.
At some point, the manager learned that:
bad decisions are punished
ambiguous outcomes land upwards
accountability sticks longer than praise
So decisions start clustering at the top.
Not by design.
By gravity.
The manager becomes the safest place for uncertainty to sit.
You’ll recognise it immediately.
Teams ask permission instead of deciding
Managers review work they don’t need to review
“Alignment” becomes a holding pattern
Everything is provisional
People are busy, but strangely cautious.
Nothing feels finished.
AI didn’t introduce this behaviour.
It just removed the excuse for it.
When options are cheap and analysis is instant, the delay becomes obvious.
If a decision still won’t move, it’s not because:
the data is missing
the scenario hasn’t been explored
the downside hasn’t been modelled
It’s because no one is willing to own the outcome.
AI didn’t slow decisions down.
It made decision avoidance visible.
It’s Confidence
Decision hoarding quietly drains teams.
Not because people want more autonomy — but because they stop trusting their judgement.
When decisions bounce back:
people hedge
people over-explain
people wait
Eventually, they stop trying to decide at all.
Why risk being wrong when someone else will step in?
The usual response is to add:
escalation paths
approval matrices
decision frameworks
That often makes things worse.
The issue isn’t process.
It’s that decision rights exist on paper but not in practice.
If someone can override you without consequence, you don’t really own the decision.
In well-run systems:
decision boundaries are boringly clear
escalation is rare and explicit
reversibility is understood, not debated
mistakes are tolerated at the right level
Managers don’t decide more.
They decide where decisions live — and then get out of the way.
That’s the job.
Letting go of decisions feels risky.
Not because teams aren’t capable.
But because the manager has to accept:
imperfect outcomes
uneven judgement
learning that isn’t tidy
Decision hoarding protects the manager.
Decision ownership develops the system.
You can’t have both.
Look at the last five decisions that really mattered.
For each one, ask:
Who should have owned this?
Who actually did?
What would have happened if the decision stayed where it belonged?
The answers are usually uncomfortable.
They’re also clarifying.
Fast organisations don’t make better decisions.
They make clearer ones.
And clarity comes from knowing:
who decides
who supports
and who lets go
The hardest part of modern management isn’t making decisions.
It’s resisting the urge to keep them.
Manager Upgrade is a private operating system for managers who want to think more clearly, make fewer decisions, and stop absorbing work that shouldn’t sit with them.
