The comment you're replying to makes clear that they're not using it on servers. I do the same thing (testing on my laptop(s), stable on my servers); you just have to commit to keeping track of what's going on and manually patch if you think that it's necessary.
I have to add that for me, testing definitely breaks stuff sometimes. Nothing beyond my ability to fix, though. Nothing beyond anyone's ability now that AI is around.
> I have to add that for me, testing definitely breaks stuff sometimes.
That's interesting, can you give an example? Asking as a curious Debian user.