a conventional CPU-based supercomputer doesn't have the necessary I/O bandwidth to do the work.
I work in HPC and the trend is towards heterogeneous architectures ( CPU+accelerators). Moore's law, power requirements and economics are dictating that trend. It's definitely a stretch to claim that you get better I/O bandwidth with GPUs. Even with PCI Gen 3, the effective bandwidth you get per CPU core is greater than that of an 'equivalent' GPU core.
Please take a look at how things work in Denmark: Video
Wow, did not see that coming.
"take out 3rd world dictators"
Can we please remove this out from our aspirations/goals? I think we already did enough harm to the world pursuing this goal.
Beware of the Turing Tar-pit in which everything is possible but nothing of interest is easy.