DES is arguably an exception to that. Twenty years after the standard was published, the EFF published designs for a machine that could brute-force a DES key in a matter of days.
But even when it was published, people were saying a 56-bit key was too small, which they aren't saying of modern cryptosystems (to my knowledge).
Yes OK, a small key size makes brute force more likely. Bruce Schneier wrote about how it's not possible to brute for 256-bit keys. Short section here http://www.schneier.com/crypto-gram-9902.html#snakeoil , there's a longer explaination saying you can't even flip all the bits in 256-bit key before the universe expires, but I can't find that at the moment. So once you hit a upper limit (256-bits), you can't brute force anything ever, so faster computers are useless. Algorithms like that get broken by new maths, which could happen at any time, you can't predict it, so you can't rely on it.
But even when it was published, people were saying a 56-bit key was too small, which they aren't saying of modern cryptosystems (to my knowledge).
http://en.wikipedia.org/wiki/Data_Encryption_Standard