The Tool That Does Everything Except the Hard Part

I use AI every day. Want to be upfront about that because what I'm about to say might sound like I'm anti-technology, and I'm not. I use it to organize ideas, draft outlines, simulate scenarios, research topics I'd otherwise spend hours digging through. It's one of the most powerful tools I've ever had access to, and I don't say that lightly coming from a world where we measured performance gains in hundredths of a second.

But something's happening underneath all that efficiency. It's the same pattern I watched play out in my own career when things got too comfortable - when friction disappeared, so did the growth.

Short track speed skating is chaos. Eight skaters, 111-meter oval, 35 mph, and the rules are "don't use your hands to shove people." Every race is a series of split-second decisions made under extreme physical and mental load - where do I position myself in the pack? When do I make a move? What's the skater behind me going to do when I shift outside?

Those decisions got sharper over time, not because someone handed me the answers. They got sharper because I got them wrong thousands of times first. I positioned too far back and got boxed in. Made a move too early and burned out with two laps left. Read another skater's pattern wrong and got taken out on the final turn. Each mistake was expensive - sometimes a medal, sometimes a season, sometimes stitches and a trip to the medical tent.

Frustration of getting it wrong over and over is what made the correct decision eventually feel automatic. My body learned it because the cost of not learning was real.

That's where my head goes when I think about what AI is doing to how we learn and work. When the tool gives you the answer before you've had a chance to struggle with the question, something important gets skipped. Not the output - the output is fine, maybe even better than what you'd have produced on your own. What gets skipped is the process of building judgment to know why that answer is right. And judgment is what saves you when the situation doesn't fit any template.

What happens to capability when friction disappears?

I've spent the last 15 years working with organizations - Fortune 500 leadership teams, founders, sales organizations, people operating under real pressure with real stakes. The pattern I see most often isn't a lack of talent or resources. It's a gradual erosion of friction that used to make people sharp.

Processes get optimized, decisions get automated, communication gets templated. People inside those systems get a little softer every quarter without anyone noticing, because the outputs still look good. Numbers are fine. Presentations are polished. Everything runs smoothly until it doesn't - until there's a crisis or an ambiguous situation or a problem without a playbook, and suddenly you need people who can think on their feet with incomplete information under pressure. Those muscles have atrophied because the environment stopped requiring them.

I watched that dynamic on the ice for 15 years. Skaters who trained in perfect conditions, with perfect coaching, with every variable managed for them - technically flawless in practice and unpredictable disasters in finals. The ones who trained in chaos, who practiced making decisions when things went sideways, who built their skills through friction rather than around it - those were the ones standing on podiums.

I'm not saying we should resist AI. That would be as productive as resisting the internet in 1998. The technology is here, it's extraordinary, and it's going to keep accelerating.

The question isn't whether to use it - it's whether we're going to let it replace the parts of us that matter. Patience, willingness to sit with a problem longer than feels comfortable before reaching for a solution. Intuition, the kind of pattern recognition that only comes from years of making decisions and living with the consequences. Ability to endure sustained discomfort without breaking. Curiosity that keeps asking why even when the surface answer is satisfying.

These are human capacities. They're built through friction. And if we outsource the friction to machines, we don't become more efficient versions of ourselves - we become more fragile versions.

I think about this a lot with younger people building their careers right now. They have access to tools I couldn't have imagined at their age, and that's exciting. But access to tools doesn't build the internal architecture to use them well. That architecture comes from struggling, from being wrong, from sitting in the discomfort of not knowing and resisting the urge to shortcut your way out of it.

The way I think about it - and I'm still working this out, at least for me - is the next decade is going to create two kinds of people. Those who use AI to eliminate friction from their lives, and those who use it to free up time and energy so they can choose harder friction on purpose.

First group will be efficient and productive and increasingly dependent on the tool. Second group will be the ones leading, creating, making the decisions that matter when the algorithm doesn't have an answer.

What I try to do now, both for myself and working with teams, is protect the struggle. Use the tool for logistical stuff, organizational work, tasks where efficiency helps. Then take the time and mental bandwidth that frees up and point it at something hard - something requiring real thinking, real patience, real discomfort.

The machine does everything except the hard part. And the hard part is still where all the growth lives.

I go deeper on this in Hard Pivot.

Subscribe to Hard Pivot

--AAO