Unless you replace the disks once in a while and do a complete file check as well once in a time you do have a lot of worry about.
These filesystems (ZFS and also BTRFS) have their own checksums, so you just need to "scrub" it, to catch the most likely issues. If you also use ECC memory, then the only thing you have to worry about is a software bug in the kernel or filesystem driver, or a hardware error in the CPU itself, causing an errant write that trashes some sensitive piece of filesystem state.
These same issues that can affect your home fileserver could also affect cloud services. There was a time when I lost some files in a OneDrive account I use at my job. Later, they mysteriously returned. I never knew what happened there, but I just have to hope there weren't any files that stayed permanently lost.
Your disks could be looking completely fine because you haven't touched the data on it for 10 years but then when the disk accesses a file on a part of the disk that has degraded it breaks down.
That's exactly why, no matter what filesystem you use, you should rely on a RAID and periodically "scrub" or consistency-check it. If using a filesystem with builtin checksums, you can do a similar operation that verifies the checksums at the filesystem level.
If you use a RAID (i.e. some level that has actual redundancy), then the negative outcome of a consistency check is usually to flag a disk as bad and in need of replacement, at which point the redundancy can be restored (RAID-6 maintains redundancy, even after 1 disk failed). In the case of a filesystem where some file fails a checksum test, your only option is to restore the file from a backup.
Both are ways of detecting and countering silent bit-rot. If you do neither, then your drives are a ticking time bomb. SSDs are generally worse on this front than HDDs.
BTW, I had a 5-disk RAID-6 of WD Black 1 TB drives that lasted over 10 years, without
any of the disks experiencing an unrecoverable read error (which you can see via SMART). Truly legendary. After I had copied the data to a newer set of drives, I accidentally dropped one of the drives about 3.5 feet onto a hardwood floor. Just for fun, I scanned the entire disk and the fall didn't damage anything in the drive mechanics that prevented it from doing a perfect scan. I wish they still made drives like that...