Well why are we still using MD5 hashes? I really want to know as I’ve heard of & seen at least one example where this is happened. This made me think that the solution might be to use multiple methods of saying it is what it says it is. I say this because all of the encryption methods I know of were developed by humans & thus have a good chance of having some sort of flaw in them. I strongly believe in the following, nothing is fool proof against a sufficiently talented fool.
My idea is to use three different hashes to significantly reduce the chances of the file being intercepted & replaced with something else. The three that I’m thinking about are MD5, SHA-1, & cyclic redundancy check. Yes I know that cyclic redundancy check is considered to be highly insecure, but I think that it would help a bit unless someone else has a better idea of what a third hash function that is better. Since they all have different mathematical functions this should reduce the chances of people faking files.
In the example I showed the files they provided only matched on the MD5 hash value with entirely different SHA-1 & CRC values. I only proposed those three because they are the only functions I know of that can determine if the files are the same. Also size is important, but in the example I showed earlier all the PDF files they had are exactly the same size. I guess this proves that one hash & the file size is not enough.