1: One does not simply turn the reactor all the way off (picture of Aragorn, err, Boromir, or who ever). There's the decay heat, several megawatts of it even after months of shutdown (google decay heat), that can't be turned off, and virtually all reactors require intervention to keep that cooled. Likewise for spent fuel pools, that boil itself out over course of a week or two. That's what happened in Fukushima - the reactors did shut down correctly but all powered decay heat removal systems failed when tsunami water flooded the basements (in which they kept the switchgear and emergency generators). The spent fuel pools were an immense problem and at least one - in the reactor building #4 where reactor was not loaded with fuel - did boil off to the point of partially uncovering the fuel, at which point they got the concrete pump to pump water in. That's Japanese, a nation of 127 millions, responding in the tsunami that killed about 16 000 people and displaced perhaps a million (excluding those displaced due to reactor). The spent fuel pool was allowed to boil itself off to the level of fuel.
AP1000 is advanced safe reactor design. It raised the no-intervention time to the whooping 72... hours.
http://en.wikipedia.org/wiki/AP1000#Design_specifications
http://www.ne.doe.gov/pdfFiles/AP1000_Plant_Description.pdf
Major safety systems are passive; they require no operator action for 72 hours after an acci-dent, and maintain core and containment cooling for a protracted time without ac power.
Which is typical hard to parse statement which translates to: after 72 hours, the cooling water reserve evaporates off, and you get yourself regular meltdown like in any other design. The protracted time without water is 72 hours. The operator action is getting a lot of water on top of 10 story building somehow. Note that most reactors in use require AC power, that's why it is so awesome AP-1000 doesn't need AC power. Also, note risk estimates at one in 10 millions years. WTF are they even doing at NRC, some sort of circle of self delusion wrt what sort of stuff happens in 10 millions years. Mankind survival is unlikely to have that sort of reliability, and that's counting the recoveries from stone age.
2: Mostly yes, i'm only speaking of radioactive pollution there.
One does not simply turn the reactor all the way off
Perhaps not, but I'd expect you'd flip a switch, and then the automated systems would shut it off, or something to that effect.
Which is typical hard to parse statement which translates to: after 72 hours, the cooling water reserve evaporates off, and you get yourself regular meltdown like in any other design.
So, set it to shut down automatically after 36 hours without operator action, and it will be fine.
can release up to 10 Chernobyls of cs-137 or so
Why didn't they have that problem at Three Mile Island?
Just a reminder that some of the old threats are still around (and hence that AI is not only something that can go hideously badly, but also some thing that could help us with the other existential risks as well):
http://blog.practicalethics.ox.ac.uk/2012/03/old-threats-never-die-they-fade-away-from-our-minds-nuclear-winter/
EDIT: as should have been made clear in that post (but wasn't!), the existential risks doesn't come from the full fledged nuclear winter directly, but from the collapse of human society and fragmentation of the species into small, vulnerable subgroups, with no guarantee that they'd survive or ever climb back to a technological society.