. . .

—“An IQ of 16000 seems absurd”—

Well the way it’s calculated is the only way it can be calculated.

I think (as Chomsky and others have suggested) that (and I have some experience testing it) that our definition of intelligence (model + forecast) today would differ from that definition of intelligence just as our two-handed nervous system differs from the eight limbs of an octopus.

In that the models we are capable of perceiving with current intelligence are limited by our capacity to act, and that at some point, the models we rely upon are not longer limited by our capacity to act, any more than our ability to measure is limited any longer by the limits of our senses, or our ability to calculate limited by our reason independent of numbers.

So we can model today what we cannot percieve with our senses directly without use of ability to gather information and reduce it to an analogy to our senses. But we can in some senses model the universe, economies and subatomic interactions.

This same ability to construct models should not have any limit that I can see other than our ability to continuously excite enough neurons to create such a model. Ergo, it should be possible. The issue is reducing cost of neural transmission and preserving the number of neurons available for learning. As far as I know that’s not difficult since we know that white matter alone does much of it.

I just don’t’ know if we’re ‘human’ any longer at that point in other than morphology.