Can you give a source for your assertion that "we have the technology to detect those"? That sounds hand wavy. I am not aware of a complex system where that kind of thing exists predictably. Especially when you're doing editing that will permanently affect the gene line, I'd say predictability is impossibly if not in theory, computationaly.
You're right, it's a little hand wavy, but we can fully sequence the DNA we extract; so I guess it depends on the error type, if there is an error that is only present in some DNA strands, that could be problematic.
