You reach for a tool because it helps. It smooths the edges, speeds things up, maybe even makes them better — just this once. You don't notice the shift until it's too late. The friction that used to signal this matters has been sandpapered away, and with it gone, so is the resistance that builds your skills, your judgment, your self-trust.
We're not talking about isolated moments of laziness. A randomized controlled trial published last month tracked 1,222 participants across a series of cognitive tasks with and without AI assistance. The results weren't subtle: just ten to fifteen minutes of AI help led to a measurable drop in independent problem-solving. The solve rate cratered from 73% in the control group to only 57%. Participants gave up faster, skipped more steps, and didn't recover once the tool was removed.
That's not a bug. It's how these systems are designed — and how we've chosen to use them.
What makes this so unsettling is the speed of agency decay. We expect dependency to be a slow burn, something we have plenty of time to resist or reframe. But the data shows a sharp decline in persistence and performance after just minutes of interaction. It's a quiet coevolution: the tool improves, we rely more, and somewhere in that reasonable sequence of small decisions, something fundamental begins to shift.
The risk arrives on tippy-toes. You justify it as efficiency, as time-saving, as cognitive offloading — all true. But it's not neutral. Every time you skip the struggle and let the algorithm fill in the blanks, you're nudging your own capacity to wrestle with ambiguity, to feel stuck and push through, to discover what you think only when the first drafts fail. That's where growth lives: in the friction between intention and execution.
The effort that is simultaneously the outcome isn't just poetic. It's physiological — neuroplastic, measurable, unavoidable. Replace the effort with an equally functional alternative and you get short-term output gains, long-term capacity erosion. It takes you through a labyrinth of small deviations, and somewhere inside that labyrinth, you stop noticing that the direction has changed.
This isn't anti-technology. It's pro-humanness. And it demands we start asking harder questions: What are we willing to trade for speed? For consistency? For good-enough outputs that require no stretch?
The answer, more often than not, is the very thing we came to use the tool for in the first place: our ability to be curious, resilient, and truly skilled.
What Friction Actually Does
We've been sold a story: friction is inefficiency. The more of it you remove, the better things get.
The evidence says otherwise — and not just as a feel-good platitude. When Mihaly Csikszentmihalyi first defined flow in the 1960s, he wasn't writing for productivity manuals. He was describing artists who would paint through hunger and fatigue, only to lose interest the moment the work was finished. Why? Because the joy — the meaning — lived in the struggle, not the artifact.
Flow requires challenge and skill in precise balance: above average on both, with clear goals and immediate feedback. Remove the challenge — strip away the friction — and you slide into boredom. Add too much without adjusting skill, and anxiety takes over. That balance isn't static; it's a negotiation you re-engage with every project, every day.
Here's the uncomfortable truth: AI tools thrive on the boredom side of that balance. They aim to deliver outcomes with minimal exertion, often landing in the zone where effort is kept just high enough to sustain the illusion of engagement — or more commonly, reduced to zero. The result? Quick wins and slow erosion.
A writer who outsources her first draft never learns what her thinking sounds like unfiltered. An analyst who asks for interpretations instead of running the tables himself never develops pattern-recognition instincts built over thousands of hours. A leader who delegates all communications misses the sharpening that separates the memorable from the merely adequate.
Growth happens at the edge of competence. AI doesn't push you there — it keeps you comfortably inside, producing passable work without ever needing to stretch.
The A-Frame for Agency Amid AI — awareness, appreciation, acceptance, accountability — isn't about saying no to tools. It's about refusing to outsource the friction that builds character and competence.
The Four Displacements
When efficiency becomes the default metric, something more subtle than simple laziness begins to take hold. Cornelia Walther calls it displacement: a quiet recalibration of values, expectations, and effort standards — each one more insidious than the last.
Let's name them. Not to shame, but to surface what often stays invisible in the rush to "get things done."
Creative displacement
Once, ambition meant chasing the perfect phrase, the exact shade of color, the tone that clicked. You'd push past the "fine" draft, because something in you refused to settle. When AI outputs are just good enough — reliable, repeatable, never remarkable — the habit of stretch atrophies. You abandon not just the sentence you didn't rewrite, but the person you might have become if you'd kept trying.
This isn't about quantity. It's about the internal compass that points toward excellence only when challenged to find it.
Social displacement
If everyone takes the path of least resistance — delegating cognitive creation rather than striving for original thought — society gravitates toward the highest common denominator. Intellectual conformity doesn't look like conformity when it wears efficiency's face.
We start believing the outputs are original, that the algorithm's suggestions reflect our own taste. Over time, we can't tell where our thinking ends and the model's begin — and worse, we stop caring enough to know.
Emotional displacement
Here's what people rarely talk about: how does AI reshaping our outputs reshape how we feel about effort? The question isn't just what the tool does to your work; it's what it does to your relationship with strain itself.
Effort becomes something to be minimized rather than understood. When frustration arises, your instinct shifts from "What's this teaching me?" to "How do I bypass it?" The friction that once signaled growth starts signaling waste — and you start avoiding both.
Aspirational displacement
Where once we recognized the satisfaction of mastery, algorithmic convenience retrains our expectations imperceptibly: toward speed, toward adequacy, away from excellence. Are we gradually moving to the highest common denominator — which may be higher in some cases, but still average for all?
The danger isn't that we settle lower. It's that we stop noticing we've settled at all.
The A-Frame for Agency Amid AI
Walther's four-step framework — Awareness, Appreciation, Acceptance, Accountability — isn't a checklist. It's a practice, repeated daily, in small ways:
Awareness
Start with your own reaching-for-AI habits. Keep a log for three days: whenever you open the tool, pause and ask — Is this about speed, or am I avoiding the effort? The inconvenience you feel before clicking "submit"? That's often the point. It's where your thinking lives.
Appreciation
Recognize the difference between AI as accelerant and AI as substitute. Processing data, formatting reports, aggregating sources — that's legitimate augmentation. But when you let the tool generate insights, structure arguments, or draft reflections that should emerge from your own experience and judgment, you're trading depth for velocity. The first is a tool. The second is a trade you haven't consciously agreed to.
Acceptance
Some friction isn't a problem to solve. It's your nervous system whispering that this task matters, that the challenge aligns with what you care about. The student who asks for hints rather than answers stays closer to the no-AI control group — their engagement, and therefore their growth, remains intact. Difficulty has its own intelligence; it tells you where growth is still available.
Accountability
Hold yourself to deliberate effort in the domains that define your distinctiveness. Your best self isn't fixed — it's an organically evolving kaleidoscope that needs challenge to flourish. No algorithm can build that.
The Irreducible Core
There's no debate: AI offers real benefits. Faster output, reduced friction, more consistency across routine tasks — these matter. But there's a category of effort for which there is no shortcut.
It's the kind that is simultaneously the outcome.
The writer who never wrestles with a sentence may never discover what they actually think. The analyst who outsources interpretation won't develop pattern recognition built over years of reading between the lines. The leader who delegates all communications misses the sharpening that separates memorable messages from the merely adequate.
This isn't an argument against delegation or collaboration. It's a warning about outsourcing the inner work — the quiet, often lonely acts of thinking, connecting, and interpreting that only humans can do well.
Csikszentmihalyi called flow "the structure of a good life." Not because it's easy, but because it requires effort. Because it demands challenge and skill in balance. Because the joy lies not in the finished product but in the process of doing — the friction, the frustration, the focus that makes you lose track of time and self.
The capacity to set meaningful aspirations, pursue them through effort, and experience the flow that comes from doing so — that's not just a performance metric. It's what makes life feel alive.
We're at risk of algorithmic aspiration adjustment — a quiet recalibration of expectations toward adequacy, disguised as convenience. But awareness is the first step toward reclaiming what's irreducible: your attention, your judgment, your effort, your voice.
Preserve them — not in spite of friction, but because of it. As the stories you tell yourself shape who you become, the narratives you sustain through deliberate effort — not delegated output — are what forge a resilient, authentic identity in the age of AI.
