One of my favorite photos that I had kept for years wouldn't open. The filename was right, and the size seemed unchanged, yet it was somehow corrupted. Even though I hadn't touched it in several months, that file was no longer the photo I had taken.

Sadly, computers don't warn you when a file changes. Maybe it's because, as far as they know, nothing has changed. But there's a real chance that your old backups or folders that sat untouched since you saved them have changed.

Your files can change without a single warning sign

Why "it still opens fine" isn't proof of anything

Picture of a WD Blue 1TB Hard Drive

When a file corrupts in the background, rather than suffering a dramatic failure, you may see a photo open with a gray, pixelated strip across the bottom. It could be a document where certain characters are replaced by weird, unexplainable symbols, or a video that may stutter, freeze, skip, or restart at a specific point.

Compared with the visible effects, the cause can feel a bit boring. Your files are stored in a physical state — magnetic or electrical — on a platter or in flash memory. Over time, that physical state can change slightly. These shifts don't trigger notifications. If even a few bits change, the file may no longer contain the data you originally saved.

This doesn't usually change a file's metadata: its name, size, and modification timestamp typically remain unchanged, because the system doesn't record the altered contents as a file change. A surface check can be reassuring — until you open the file.

When this data degradation happens, you should understand that the storage device still works, but that doesn't mean your data is intact. No folder icon tells you the difference.

Your file system isn't watching your files the way you think

It protects the drive's structure — not what's inside your files

NTFS file system

Drive health, file system health, and file integrity are often lumped together, but separating them helps. Modern operating systems monitor drive and file system health, but they don't check file integrity, which is often the element you care most about.

Windows, Mac, and Linux file systems (NTFS, APFS, and ext4) have a well-organized internal bookkeeping architecture. They know where files are stored and what folders are associated with specific files. However, they don't check whether a file's contents still match what you originally saved.

Two notable exceptions are Btrfs and ZFS. These keep a fingerprint of the data you store and recheck it when a file is read to catch the kind of decay I'm referring to.

Filesystem

Verifies file content?

NTFS / APFS / ext4

No

Btrfs / ZFS

Yes

If you already own a NAS, there's a good chance it's running Btrfs or ZFS by default, so you might be more protected than you realize without having done anything.

I don't expect you to reformat your drives over this; the next part is more important than switching file systems.

Verify your files with a checksum

A checksum makes silent corruption easy to spot

A checksum is a file's fingerprint. If you run the same file through a checksum several times, you'll get the same result. However, the fingerprint is totally altered by changing even the tiniest bit in a file. This is why it's the most reliable way to catch file corruption your system won't notice.

Today, the standard I trust is SHA-256. MD5 is faster to compute and is still adequate for detecting accidental corruption, although SHA-256 is the better choice overall. On Windows, the Get-FileHash command generates a checksum for a file; on Linux and Mac, you can use shasum -a 256. However, several GUI tools make this process easier: OpenHashTab on Windows, Hash on macOS, and GtkHash on Linux. With these free checksum GUI tools, you can drag a folder in, generate fingerprints for everything inside, and save that list somewhere safe.

If you generate a checksum and never check again, all you have is a fingerprint sitting in a drawer. I set a recurring reminder to check every few months.

I recommend you do this for the files you rarely open, because they stand the greatest risk of bit rot. Once the checksum returns a value different from what you recorded in the past, it's an instant indication that the file is damaged, and restoring the file from a known-good copy is usually your best option.

A backup isn't proof your files are safe

While a backup makes it harder to lose a file, it doesn't mean bad copies can't be faithfully preserved. Most backup software simply copies data bit-for-bit, exactly as it finds it. This is why damaged originals also mean damaged backups once the sync is done. All you end up with is two identical, broken files.

I now keep two copies of each important file in different places. From time to time, I go back and open what's important to me. I no longer trust that once it's backed up, it's fine because, if you've never opened that backup, at best you're guessing it's OK.

Right now, pick your most irreplaceable files, confirm that they exist in different places, and open each one to verify that it isn't damaged. It's a simple habit that makes a bit rot much less likely to catch you by surprise.