You need the small subset of people who actually understand how the software works to fix the bugs, lest you just end up playing whack-a-mole and introducing new bugs.

Ideally, it wouldn't be like this, but we don't live in an ideal world, and there's always some unintended consquence when you start pulling the string on anything nontrivial. And there's never sufficient automated testing, especially for the things that haven't been a problem yet.