It's not about the technology

7 April 2026

I've been watching two people swap positions on AI over the last eighteen months. Both are experienced technical leaders. Both work in the same business, navigating the same pressures, looking at the same industry. One started as a sceptic and has moved, gradually and honestly, into something closer to optimism. The other started as an enthusiast and has recently become worried.

Neither of them changed their mind because the technology changed.

The sceptic spent the better part of a year watching AI fail on the kinds of problems he actually cared about: complexity, existing codebases, enterprise-grade thinking. He wasn't wrong. The limitations were real. What shifted wasn't the argument. It was the moment he put a genuinely hard problem in front of a model, and the model not only solved it but found things he hadn't thought to look for. His concerns about the technology didn't disappear. But something about his relationship to it did.

The enthusiast had been energised by the possibilities. He'd moved fast, seen ideas become MVPs quickly, felt the pull of what was coming. Then he got close enough to see the other side. Real codebases left in a mess by developers who had leaned on tools they didn't fully understand. He cleaned it up. And then, like many businesses right now, the numbers came in harder than expected.

When the future feels less certain, things that once felt exciting start to feel threatening.

What strikes me about both of these trajectories is that the technology itself is almost incidental. The sceptic's turning point wasn't a new benchmark or a product announcement. It was a personal experience that let him test his own value against the machine and find it wasn't diminished. The enthusiast's shift wasn't triggered by a technical failure. It was triggered by proximity to the consequences of inexperience, and by results that quietly changed what the future looked like.

I've felt both sides of this myself.

People's relationship to AI is a mirror of their relationship to uncertainty.

I see this pattern everywhere I look now. There are people so close to the news cycle that the speed of change has become its own kind of overwhelm. Every week brings something new that resets the map. There are others who have decided to wait, to let the noise settle, to see what actually sticks. There are developers at the bleeding edge, genuinely energised by being early in something unfathomable. And there are experienced engineers who find the whole thing threatening in ways they haven't quite named yet. They carry thirty years of hard-won knowledge and are quietly wondering whether any of it still counts.

The catch-22 is uncomfortable. The more experience you have, the more you have to lose, and the harder it is to approach something new without that weight. The less experience you have, the more freely you can experiment, but the more damage you can inadvertently do.

We've been here before, in a sense. The Luddites weren't wrong that the looms would change everything. They were wrong about whether that meant the end. Humanity has come through waves of technological disruption, and each time the fears were real and the adaptation happened anyway. But I'm not sure that history fully settles the question. Materially, we seem to come out ahead. Whether we come out better in other ways is less obvious to me.

I don't have an answer to that. But I notice it changes how I hold the question.

What I do know is that the people I see navigating this well aren't the ones who have figured out the technology. They're the ones who have a stable enough relationship with their own uncertainty that they can stay curious without being destabilised. They can hold doubt and possibility at the same time without needing to resolve it prematurely in either direction.

The sceptic didn't become an optimist because the models got better. He became one because he found a way to keep experimenting until the evidence he needed showed up. That took time, and a willingness to be wrong, and enough room to fail without it meaning something bigger about who he was.

That's not a technology problem.

That's a human one. It always has been.