That proof is very confused. The ability to detect duplication in no way proves the files are unencrypted (indeed this should be obvious from the fact that the

That proof is very confused. The ability to detect duplication in no way proves the files are unencrypted (indeed this should be obvious from the fact that there is only negligible network traffic to confirm a duplicate! The bits can't be compared if they're not transmitted.) It's the ability to serve deduplicated files that brings the service into question. Yet I wouldn't be surprised if there exists an asymmetric encryption method which permits decryption with one of several private keys – if so, secure deduplication is trivial: confirm the duplicate using a hash or comparing public-key encrypted versions; re-encrypt using both original and duplicate keys. (And let's not even forget the ability to reset a forgotten password…)