No. MD5 is not encryption (though it may be used as part of some encryption algorithms), it is a one way hash function. Much of the original data is actually “lost” as part of the transformation.
Think about this: An MD5 is always 128 bits long. That means that there are 2128 possible MD5 hashes. That is a reasonably large number, and yet it is most definitely finite. And yet, there are an infinite number of possible inputs to a given hash function (and most of them contain more than 128 bits, or a measly 16 bytes). So there are actually an infinite number of possibilities for data that would hash to the same value. The thing that makes hashes interesting is that it is incredibly difficult to find two pieces of data that hash to the same value, and the chances of it happening by accident are almost 0.
A simple example for a (very insecure) hash function (and this illustrates the general idea of it being one-way) would be to take all of the bits of a piece of data, and treat it as a large number. Next, perform integer division using some large (probably prime) number n and take the remainder (see: Modulus). You will be left with some number between 0 and n. If you were to perform the same calculation again (any time, on any computer, anywhere), using the exact same string, it will come up with the same value. And yet, there is no way to find out what the original value was, since there are an infinite number of numbers that have that exact remainder, when divided by n.
That said, MD5 has been found to have some weaknesses, such that with some complex mathematics, it may be possible to find a collision without trying out 2128 possible input strings. And the fact that most passwords are short, and people often use common values (like “password” or “secret”) means that in some cases, you can make a reasonably good guess at someone’s password by Googling for the hash or using a Rainbow table. That is one reason why you should always “salt” hashed passwords, so that two identical values, when hashed, will not hash to the same value.
Once a piece of data has been run through a hash function, there is no going back.
You can’t – in theory. The whole point of a hash is that it’s one way only. This means that if someone manages to get the list of hashes, they still can’t get your password. Additionally it means that even if someone uses the same password on multiple sites (yes, we all know we shouldn’t, but…) anyone with access to the database of site A won’t be able to use the user’s password on site B.
The fact that MD5 is a hash also means it loses information. For any given MD5 hash, if you allow passwords of arbitrary length there could be multiple passwords which produce the same hash. For a good hash it would be computationally infeasible to find them beyond a pretty trivial maximum length, but it means there’s no guarantee that if you find a password which has the target hash, it’s definitely the original password. It’s astronomically unlikely that you’d see two ASCII-only, reasonable-length passwords that have the same MD5 hash, but it’s not impossible.
MD5 is a bad hash to use for passwords:
It’s fast, which means if you have a “target” hash, it’s cheap to try lots of passwords and see whether you can find one which hashes to that target. Salting doesn’t help with that scenario, but it helps to make it more expensive to try to find a password matching any one of multiple hashes using different salts.
I believe it has known flaws which make it easier to find collisions, although finding collisions within printable text (rather than arbitrary binary data) would at least be harder.
I’m not a security expert, so won’t make a concrete recommendation beyond “Don’t roll your own authentication system.” Find one from a reputable supplier, and use that. Both the design and implementation of security systems is a tricky business.