Way back when, my Dad always told me to have at least three copies of everything, because one backup is no backup. So I have a 1TB Storage Box at Hetzner (€4/month, in a different datacenter than my VPSes) I push backups to with Restic. I also backup to a local machine (also with restic), and to my Dad’s NAS (you guessed, with restic), and to my old Debian server until I decommission it (also with restic).
All legs have different retention settings, based on how much space I have at a particular backup site.
Most crawlers don’t care - they’re purely downloading. Most garbage will also be filtered out before training, and generating plausible garbage is expensive - not really worth the trouble. What you can do with the garbage, is that you can place unique identifiers in the links you serve, so when a crawler that you can’t recognize otherwise comes visiting a poisoned URL, you can firewall it off. This works because the crawlers use a queue: they get a list of URLs to visit, they download them, and later on, the URLs from those get extracted, and crawlers will visit those URLs.
So if you manage to trap a bot, and overwhelm its queue with poisoned URLs, you can also catch bots that you wouldn’t be able to easily identify otherwise. And that is why serving an endless maze of garbage is so very useful. Because it lets you catch stuff like headless chromes operating through residential proxies: because those too, follow a queue.