Who reviews 20,000 lines?
It's probably great, and helps you as a purist in terms of coding, to have an understanding of the code that AI is writing. But the fact is that AI is going to keep writing way more code than humans can review.
There's a difference between building something that works and then building something for production. When we start getting into the production level, maybe that makes sense, to have a fine understanding of everything that's going on.
But in reality, there will be way too much for you to account for. Before, when we had humans reviewing code, there wasn't a lot of code to review, because another human wrote it. But right now AI in five minutes spit out a 10,000, 20,000 code repo. If you're going to take your time going through that, the question is: of what value is that to this space and world that we're in right now?
Again, there's an argument for: at some point somebody needs to review the mess that has been going on. But I think the jury is still out on that.
Said out loud in a team meeting, 17 April 2026. Trimmed for filler; not rewritten.