... Ok everyone uses it (the exploit) and difficulty rises by 1000x.
That is everyone except the ones with non-programmable asic chips.
True but there will be a very large incentive for an ASIC developer to release a chip which takes the crytographic flaw into effect. Until that happens you could see (hypothetically) a period of time where the average GPU can out hash a high end ASIC.
I'm doubtful that an ASIC developer would release a chip that makes use of a broken hash function. The chance would be too high that Bitcoin would switch to a new algorithm, rendering their new (and old) chips worthless.
Really so it would be better to see their product instantly become worthless, sales plummet to zero overnight, and all miners revert to using (cryoptographic flaw enhanced) GPU? That is better than a (essentially 0.0%) chance that Bitcoin will switch to a new algorithm. Note the clumsiness of the saying "Bitcoin will switch". Is Bitcoin a person? a corporation? Who exactly would switch? EVERYONE? Nobody would decide to keep using "original bitcoin". Imagine the confusion and chaos if for example MtGox supported original Bitcoin but bitpay supported "new bitcoin" and there were two incompatible forks of the reference client each supporting one version of the protocol and users had coins on both versions with initial balances at the point the blockchain permenently and irrevocably split. Think that chaos might be bad for Bitcoin? Think that might be worse than the non-problem of miners calculating hashes faster?
The hashing algorithm isn't going to change. Anyone who understands the decentralized consensus that is Bitcoin understands that. A flaw which allows someone to hash 1000x, 1,000,000x, hell 50 quadrillion times faster presents absolutely no risk to Bitcoin. It simply means everyone will end up using the "flaw" (either back on CPU/GPU/FPGA or on some future ASICs) and difficulty will increase by an equivalent amount.
When weaknesses like that are found, the world generally takes it as a warning sign. It means that future weaknesses are very likely to be coming, and worse. In that scenario,
everyone would have a strong incentive to change. But we'd also have lots of time to do it.
Look at MD4. It is considered to be totally broken, and should not be used for anything, ever. But really, none of the attacks on MD4 are even remotely useful for mining. If we were using MD4 for mining, we would have years, probably decades really, to engineer a switch.
And a switch can be relatively painless. As an example, we could all agree that ~2 years is plenty of time for everyone to upgrade their software, and ~4 years is enough to upgrade hardware. That would mean that starting with block 3,150,000 the network would accept blocks with valid hashes under either SHA or the new algorithm, whatever we decide that may be. And starting with block 4,200,000 the network would stop accepting blocks with SHA.
