If we're talking about bitrot protection, I'm pretty sure S3 would use some form of checksum (such as crc32 or xxhash) on each internal block to facilitate the Reed-Solomon process.

If it's verifying whether if it's the same file, you can use the Etag header which is computed server side by S3. Although I don't like this design as it ossifies the checksum algorithm.