That was how I started reflecting on it (again, the Castle Heterodyne model is very useful).
One thing I wanted to be careful with in that whole ecosystem was that there was a qualitative difference that hypergenius brought to the mix. Hypergeniuses are not just really really smart geniuses – I think I posited (or agreed with whoever suggested it in the collaboration) even during the game (and certainly afterward) that their perceptions and unconscious understanding of things down to the quantum level allowed them to actually Change How Things Work. They weren’t just discovering new shit – they were making new shit, the observer changing the observed, which is usually why their inventions are irreproduceable.
That doesn’t mean, though, there can’t be gradations, and a lot is dependent on more human qualities of discipline, focus, vision, process management, and control. It’s not just power, but the ability to apply it usefully to greater ends that defines the hypergeniuses that newspapers write about and governments develop programs like Antibody for (and then, probably, spend a lot of money or a couple of bullets taking off the board completely or for the most part – it’s hard to keep a Byron Quill or Achilles Chin restrained, significantly harder than a Sidorov.
I suspect one reason why hypergeniuses are so unstable is that their ability to tweak reality inevitably turns inward. “If I were even smarter / more disciplined / more creative /. less alcoholic, I would achieve even greater things / be greater than Achilles Cheka / get out of this cell / rule the world!” Unless they have given at least some study to neurobiology and the like, they are as likely to do harm as good (if not both). “See, now I can see at a atomic level and I have suppressed my fear reflex! The world is my oyster!” (to be shortly followed by a lab-shattering ka-boom).