"You didn't write the code, so you don't truly understand it." Is a common argument against AI coding tools.

But it doesn't hold up.

I just read another post making this case. It sounds reasonable on the surface, but it ignores something every experienced engineer knows.

Most of the code you work with, you didn't write.

Really, truly think about that for a moment. When I do, I’d estimate that for many of the codebases I've worked on less than 1% was mine. The rest? Written by hundreds, sometimes thousands of engineers who worked on the codebase before, or alongside me. My job was still to understand it, debug it, ship on top of it.

That's not the exception. That's the pattern I've seen time and time again across 30 years of software development.

Reading and reviewing code has always been a core skill. AI-assisted coding doesn't change that.

So the next time someone makes this argument, ask yourself: how much of the codebase you work on did you actually write?

For most of us, the answer is a number that'll surprise you.