Will I Still Matter? The Deeper Question Behind AI Adoption
AI adoption is a work-design question: preserve human autonomy, authority, and agency while creating capacity, capability, and clarity.
Organizations are racing to answer a seemingly straightforward question:
What work can we give to AI?
It’s an important question. But it may be the wrong place to start.
Much of the conversation about employee resistance to AI centers on fear: fear of job loss, unfamiliar technology, or simply change.
Those fears are real. But I don’t think they fully explain what we’re seeing.
There may be something more fundamental underneath AI resistance:
People don’t just want to keep their jobs. They want to matter in them.
What are employees actually afraid of losing?
Research into motivation, job design, and employee experience has long pointed to the importance of control, judgment, mastery, and meaningful contribution at work.
I think about those needs as the 3 A’s of human work:
Autonomy: I have meaningful discretion over how I accomplish my work.
Authority: My expertise and judgment carry weight.
Agency: My choices influence outcomes, and I can see how my work creates value.
Now consider what can happen when AI enters a workflow.
The system recommends the answer.
The system determines the process.
The system drafts the work.
The system evaluates the result.
At some point, an employee may reasonably wonder:
What exactly am I here to do?
That’s not necessarily resistance to technology. It may be a response to a perceived loss of professional value.
The wrong goal: maximum automation
This distinction matters because organizations can inadvertently automate the parts of work that make a job worth doing.
Automate 70% of someone’s responsibilities—administration, searching, reconciling, formatting, scheduling—and you may dramatically improve their job.
Automate 20% of someone else’s—but make that 20% the judgment, relationships, creativity, or ownership—and you may make their job substantially worse.
The percentage of work automated matters less than which work is automated.
The goal shouldn’t be maximum automation.
It should be maximum human leverage.
The other side of the equation
If AI takes work away, it should give something meaningful back.
I think of those benefits as the 3 C’s:
Capacity: Remove low-value work and give people time and attention back.
Capability: Help people analyze, create, decide, and accomplish more.
Clarity: Reduce noise and friction so people can focus on what actually matters.
Put the two together:
Use AI to increase Capacity, Capability, and Clarity without unnecessarily diminishing human Autonomy, Authority, and Agency.
That’s a very different objective from simply “driving AI adoption.”
Resistance is information
This also changes how leaders should respond when employees resist an AI-enabled workflow.
The instinct is often to reach for change management: more communication, more training, more champions.
Sometimes that’s exactly what’s needed.
But first, diagnose the resistance.
Which A feels threatened?
Is AI reducing Autonomy by dictating how work gets done?
Is it reducing Authority by displacing professional judgment?
Is it reducing Agency by separating employees from the outcomes they once owned?
Then ask the reciprocal question:
Which C are we actually creating?
Are employees gaining meaningful capacity, capability, or clarity?
If we’re taking away an A without creating a meaningful C, we may not have an adoption problem.
We may have a work-design problem.
Designing the human-AI boundary
The most important workforce question isn’t simply:
Human or AI?
It’s:
Where should the boundary between them sit?
Some work should belong to machines. Some should remain distinctly human. And much of the opportunity sits between the two—combining machine capability with human judgment.
Finding that boundary requires more than technology strategy. It requires understanding workflows, decision rights, organizational design, incentives, employee experience, and the economics of work.
The organizations that win won’t necessarily automate the most.
They’ll redesign work most intelligently—removing the friction that keeps people from doing their best work while preserving the autonomy, authority, and agency that make their contribution valuable.
That’s the question I’m increasingly interested in:
How do we redesign work around what technology can do without losing sight of what people need from work?
Because the future of work isn’t just about what AI can do.
It’s about what humans get to do next.