Mind…fricken…blown! Wow!
I was watching the above YouTube video entitled The AI Rollout is Here – And It’s Messy by the Financial Times, when at one point within it one of the people being interviewed said the following statement (at the 7:47 minute mark).
The difference this time is the inherent capability of the systems is so much greater that you need a lot of training to be able to fundamentally change the way you work…
To put this more specifically, the potential of AI is massively being underused because people don’t know how to effectively use it yet because it requires a change of mindset in the way they do their work.
To see if I was truly perceiving and understanding this at the deeper level I thought I was, I asked ChatGPT the following question. It’s answer pretty much mirrored my intuitions, with the key takeaway below.
Most people can’t use AI well yet because the psychological, cognitive, and identity shifts required are bigger and more uncomfortable than the technical learning curve.
NOLLIND
How are most people not achieving the true potential capabilities of using AI right now because they don’t have the training yet to change the way they work with it?
CHATGPT
Most people under-use AI right now not because the tools are weak, but because their mental models for how to work haven’t updated. The constraint is cognitive, not technical. Let’s break the gap down directly and challenge a few assumptions along the way.
1. People Still Treat AI As a Faster Search Engine, Not a Cognitive Partner
Most users approach AI with “give me an answer” questions.
That’s a holdover from Google-era thinking: query → result.
AI requires a dialogical approach—iteration, refinement, co-construction.
But this feels unnatural because people aren’t taught to think in loops. They think in requests.
Counterpoint: Even when people are told to iterate, the habit rarely sticks because they still expect linearity. They don’t yet have an embodied sense of “thinking with” rather than “asking from.”
2. They Haven’t Shifted From Task Thinking to System Thinking
AI shines when used to build frameworks, workflows, drafts, rewrites, and simulations—not one-off answers.
But most workers define their job as executing discrete tasks. Their identity is wrapped around visible effort, not invisible systems design.
So they don’t realize the real leverage is in changing the shape of the work, not speeding up the same old shape.
Challenge: Even tech-savvy users unconsciously protect their existing workflows because those workflows reinforce their sense of competence.
3. They Lack Prompt Literacy—but the Deeper Issue Is Conceptual Literacy
The common advice focuses on crafting better prompts, but that misses the point.
The core skill isn’t prompt writing; it’s articulating thinking—assumptions, constraints, goals, criteria.
Most people aren’t trained to externalize their reasoning, so they give AI vague inputs and get vague outputs.
The overlooked truth: AI exposes gaps in the user’s own clarity more than it exposes gaps in the model.
4. People Don’t yet Think in Terms of Delegation
AI is essentially a general-purpose junior collaborator.
But most people have no training in delegation, creative direction, or iterative supervision.
It’s the same reason many managers struggle: they’re used to doing the work, not orchestrating it.
Implication: The shift to AI requires a shift to meta-work—thinking about the work at a higher level. Most people’s schooling never trained that capability.
5. Emotional Resistance Blocks High-Leverage Use
This is rarely acknowledged, but it’s one of the most powerful forces.
- Fear of losing control
- Fear of feeling stupid
- Fear of obsolescence
- Fear of exploring unfamiliar territory
- Fear of seeing one’s own limitations reflected back
AI doesn’t just accelerate work—it confronts you with your habits.
Most users unconsciously avoid modes of interaction where they feel exposed or incompetent.
6. Vertical Development Matters
At a deeper level, using AI well demands the very capacities you explore in your own work:
- managing uncertainty
- questioning assumptions
- holding multiple perspectives
- revising mental models
- co-creating with something partially unpredictable
Most of the population is still operating from earlier mindsets where authority, correctness, and clear rules dominate.
AI dissolves fixed rules. That alone disorients people.
From that perspective, AI isn’t just a tool—it’s a developmental pressure.
7. They Haven’t Unlearned the Idea That “work = Output”
High-level AI use is about shaping the generator, not generating the thing yourself.
It’s designer thinking, not producer thinking.
But internalized work identity is slow to change.
If you look across all these points, the pattern is straightforward:
Most people can’t use AI well yet because the psychological, cognitive, and identity shifts required are bigger and more uncomfortable than the technical learning curve.



