ZFS is a filesystem, the idea that it has something to do with creating backups is just a moron flag on the speaker.
Who says ZFS is a backup? Can you cite these sources, because most of the knowledgeable people (including myself) who run ZFS treat it as a RAID, not a backup. And a RAID that happens to have some very handy and useful features like snapshots, a simple integrated and open command line and so on. The only place I consider ZFS as a backup is on my backup array... the one my primary replicates to every four hours (though could easily be every 15 minutes if I chose that... or heck I could script it for rolling backups if I really wanted to get fancy). And that's just my "first tier" backup.
If you're not a hardcore filesystem nerd who already checks their filesystem stats by hand on the cli frequently, then it is just another filesystem and believing it makes your data safer makes your data less safe. A false sense of a security is not harmless.
Every default installation of ZFS I've done in the last couple of years has had auto-snapshotting and auto-scrubbing as a cron job. How often you run the scrub is entirely up to you and you can control the snapshots easily with filesystem settings.
If you have backups, then you restore the corrupted file. It is simple. And even if you were a filesystem nerd, faster than fucking with the filesystem to try to recover whatever had bit rot. Just replace the file from the backup.
Completely true... but how do you know what file is corrupted and when it got corrupted? That's the challenge of course... and bear in mind depending on your backup cycles you may not actually have a good copy of the file in your backups. Nice thing about ZFS is the auto-scrub will identify the corrupted file at a minimum and allow me to revert to a previous snapshot of that particular file easily. I scrub monthly, so if a corrupted file is found it'll be no more than a month old since the scrub walks the entire pool (so verified monthly). Since I keep up to yearly snapshots on my array I can go back up to a year in just snapshots and simply copy the file over the corrupted one without having to resort to my onsite or offsite backups.
ZFS does have the capability to also fix a corrupted file, but its use cases and caveats are beyond the scope of my post. Even at its simplest, ZFS can at least identify a corrupted file even on a single disk array due to checksums.