As far as I can tell, Watson is obviously right.
That said, we have every reason to believe that the intelligence gap is solvable. The difference in performance between American blacks and those still in Africa suggests that more than half of the problem is environmental and can be tackled via relatively simple measures like micronutrient supplementation. And by the time we are done implementing that, it's likely that germline engineering and/or "smart pills" will be sophisticated enough to handle any genetic component of the problem that may remain. So there is no technical reason for doom and gloom. And the course of action described here shouldn't even be that expensive.
The problem is political will. The subject of racial differences in intelligence is too politically charged domestically for any US politician to be able to lead such an effort anytime soon. My guess is that China's leaders aren't similarly constrained, but they're just interested in securing natural resources from the continent and don't care for "uplift". So the best hope may be philanthropists like Bill Gates...
I interpret Eliezer to be saying that the Kolmogorov complexity of the human genome is roughly 25MB -- the absolute smallest computer program that could output a viable human genome would be about that size. But this minimal program would use a ridiculous number of esoteric tricks to make itself that small. You'd have to multiply that number by a large factor (representing how compressible, in principle, modern applications are) to make a comparison to hard drive bits as they are actually used.