Every generation gets its own version of this fight.
Desktop publishing wasn't real design. Photoshop wasn't real photography. Digital audio wasn't real music. The internet wasn't real business.
Now it's AI.
I've listened to the arguments. Some of the concerns are legitimate. Some of them are people protecting a way of working they're comfortable with and calling it principle. After a few years of actually building with these tools — not reading about them, building with them — I've landed somewhere simple:
I'm not interested in suffering for the sake of tradition.
If a tool helps me think more clearly, solve a problem faster, build a better system, or learn something I didn't know yesterday, I'm going to use it. That's the whole position. That's exactly what AI does.
AI doesn't do the work
This is the part that gets lost.
AI doesn't decide what matters. It doesn't decide which problem is worth solving. It doesn't know what makes a good website, a clean workflow, a song worth finishing, or a strategy that actually holds up. People do.
The value was never in the typing. It's in the judgment. The experience. Knowing when something is wrong before anyone else notices it. AI can generate a hundred options in a second. It still can't tell you which one is any good. That's taste, and taste doesn't come in a model.
In my own work, AI helps with research, planning, documentation, troubleshooting, writing, code, and rapid prototyping. It moves an idea from "in my head" to "running on a screen" faster than I could ten years ago. But the thinking is still mine. The decisions are still mine. And when something ships broken, that's on me too, not the tool.
The difference isn't access. It's application.
A lot of people talk about AI like everyone with the same tools will produce the same results. That has never been true of anything.
Two people own the same camera and take completely different photographs. Two people own the same guitar and write completely different songs. Two people open the same AI model and build completely different things.
The tool isn't the differentiator. The person using it is.
Honestly, the more I work with this stuff, the more convinced I am that expertise matters more now, not less. Knowing what to ask, what to ignore, what to double-check, and what's actually worth building — that's the skill. It always was. AI just raised the ceiling for the people who have it, and exposed the ones who were faking it.
What I'm actually interested in
Not the hype. The practical stuff:
- Building local AI agents that do real work
- Automating the repetitive tasks that eat a week
- Connecting systems that were never designed to talk to each other
- Using AI to manage websites, reporting, and content pipelines
- Poking at robotics and media production
- Speeding up design, music, writing, and development
Not because AI is magic. Because time is finite. If a tool gives me back ten hours a week, I get to spend those ten hours building better things. That's the entire trade, and it's a good one.
The research keeps pointing the same way
You don't have to take my word for it. The studies coming out keep landing on the same handful of findings, over and over.
AI lifts productivity the most for the people who actively steer it and check its work, not the ones who paste in a prompt and take whatever comes back. It helps creativity as a collaborator, not a replacement for judgment. And it doesn't turn a novice into an expert, it makes a capable person faster.
The through-line is consistent: AI is changing how the work gets organized, not erasing the need for someone who knows what they're doing.
The future isn't more hype
I don't think the future is giant promises and another wave of AI marketing. I think it's people quietly getting good at using these tools well. Not replacing human judgment, extending it. Not removing creativity, amplifying it.
The question was never whether AI exists. The question is whether you're willing to learn to work with it.
I am. That's why you'll see AI all over this site, not as a stand-in for people, but as a tool that, in the right hands, helps people do more than they could before.
So, how are you using it?
That's the part I actually care about.
I'm building this stuff in the open on purpose, because the most interesting conversations right now aren't the hot takes about AI. They're between the people quietly figuring out how to use it. The ones automating their own grunt work. Wiring up agents. Breaking things, fixing them, and learning where the real edges are.
If that's you, I want to compare notes. What have you built? What actually worked? What turned out to be a dead end? Reach out, drop a comment, send me what you're working on. Nobody owes the internet an apology for using better tools. The interesting question is what we're building with them, and I'd rather figure that out next to other people doing the same.