It's good to see the Edge contributors coming to a consensus, lol.
I think Sam overlooks a few things here.
"Yes, other fields pose extraordinary risks—but the difference between AGI and something like synthetic biology is that, in the latter, the most dangerous innovations (such as germline mutation) are not the most tempting, commercially or ethically. With AGI the most powerful methods (such as recursive self-improvement) are precisely those that entail the most risk."
vs
How recursively self-improving organic robots will modify their own
source code and bootstrap our way to full-spectrum superintelligence
The Biointelligence Explosion