"AI" is a word like "electricity" or "internet." Broad enough to mean anything.
Yes, the internet can be bad if used for crime. Electricity can be bad if used to hurt people. AI can be bad too. It can also be useful. The technology doesn't decide. We do.
Content that spreads picks a side. Pro or against. No middle ground. That's not because the middle ground doesn't exist. It's because anger and division drive engagement. The loudest voices benefit from your attention, not your understanding.
I want to offer a third option. Not pro-AI. Not anti-AI. Just: try it, and decide for yourself.
I've Seen This Before
I'm a programmer and game developer. That's the world I know. I won't pretend to have answers beyond it.
My first encounter with programming was at a flea market in Mexico City. I was maybe 13. My dad and I were walking through rows of pirated CDs when one caught my eye: Borland C++. I asked the seller what it was. He said it was a program to create other programs.
That blew my mind. But I honestly don't recall understanding what he meant.
My dad bought me the CD. I went home, installed it, opened it, and—nothing. I had no idea what to do. I didn't know what a programming language was. I didn't know about syntax or compiling. I just sat there, confused. Not disappointed. Just not ready.
Years later, I took a workshop. HTML, JavaScript, Flash. I rode the metro to get there, and on the way I'd pass bookstalls selling computer magazines. One day I found a C++ book in Spanish. I read it without ever turning on a computer. Just trying to understand how someone could describe a concept—like a vehicle—in a way a machine could understand.
Then came Java at university. Then C. Then C#. Each layer built on the last. Each one made the next one easier to learn.
At some point, I discovered XNA and started making games. Later, Unity. I wrote shaders by hand—I loved designing them, but the typing was just a means to an end. Then Unity introduced Shader Graph, and I never went back. Nobody called me a fake shader programmer.
The same thing happened with game engines. People used to insist real developers wrote everything in C++ from scratch. Now engines are standard. We forgot the argument. Kind of. Now the argument is which engine is best.
I've watched this pattern repeat my whole career. Tribal resistance, then adoption, then we move on.
What Changed For Me
I used ChatGPT for a while. I went to meetups in Austin, San Francisco, New York. People talked about Cursor and Windsurf. It didn't click. I couldn't tell what the AI was doing versus what I was supposed to do.
Then I tried Claude. I liked the style, the design. I learned about Claude Code.
I still wasn't sure what I could do with it. But I'd been learning threejs, so I started there. I asked Claude to help me build prototypes. The speed surprised me. I kept talking to it, kept building.
And that's when thing shifted and finally clicked.
I had ideas I'd been carrying for years. Gameplay concepts. App ideas. System designs. Things I thought were good but never had time to build properly. With Claude Code, I could prototype them fast. Test them and see if they were worth pursuing. Most didn't but now I could take them out of my system faster.
That's the real value. Not building faster—though that's part of it. The real value is knowing what not to build. Ideas I might have spent years on, I could kill in days.
What the Tool Does and Doesn't Do
The way I see it, programming has two parts.
One part is thinking: solving problems, designing systems, deciding what to build and why. The other part is typing: syntax, APIs, boilerplate, looking things up in documentation.
I've outsourced the typing to AI. The thinking is still mine, and the AI.
This isn't new. Auto-complete changed how I wrote Java. IDEs changed how I navigated code. Shader Graph changed how I made shaders. Each tool shifted what I spent my time on.
Claude Code is the same pattern for me. Just at a bigger scale.
The parts I enjoy—figuring out the problem, choosing the approach, making decisions—those are still mine. If anything, I do more of that now because I spend less time on the rest.
If you love typing code, keep typing. If you love learning language syntax, keep learning. Use AI for the parts you don't enjoy. Or don't use it at all. That's fine too.
The question isn't whether the tool is good or bad. The question is whether it helps you do what you care about.
On Art
I love looking at art. Museums, Twitter, books. I'm not an artist, so when I see something beautiful, I wonder: how did they make this? What were they thinking?
AI-generated art(maybe we should not use the word Art yet?) doesn't hit the same way for me. When I look at it, my brain just thinks: noise plus prompt. Yes, you can trick me, but that's more on you than on me. Humas, we are easy to trick.
Something similar happened with CGI. When I was younger, I'd watch movies and wonder at impossible things. Then I learned how CGI worked, and some of the magic faded. It's still impressive—but different.
Maybe one day I'll connect with AI art. Today I don't. That's just my preference, not a universal truth.
We get to choose what we consume. That's a kind of power. But the choice should come from what we actually want—not from what someone told us we should want.
The Real Question
I'm not here to tell you AI is good or to tell you it's bad.
I'm here to say: don't think too much about good or bad. Think about what can positively affect you, your life, and the people and environment around you.
That might mean using AI. It might mean avoiding it entirely.
Here's something that helps me: I ask who benefits from each side of the debate. The people shouting loudest—do they gain something from my agreement? From my outrage? From my attention?
That doesn't tell me what to believe. But it tells me to be careful about why I believe it.
Close
Life is short.
If something might help you live better, at least give it a try. Form your own opinion from experience, not from inheritance.
You can't control the shouting match. You can't make the algorithm reward nuance. You can't stop people from picking teams.
But you can control what you do, what to try and what to choose.
That's the third option. The debate is jurassic. Your choice doesn't have to be.