I tried to do some of that in my comment. I didn’t bother passing their ITT (think it’s a waste of time), but tried to make the case that (what I take to be) their verbally stated goals clearly are not being matched by their actions[1].
I did state my arguments in rational-ese and EA-adjacent language but would be happy to see similar presentations in more Western Buddhist-flavored language if they think on the fence MAPLErs or would-be future residents would find it helpful.
imo there are ~4 options to argue against my comment:
a. Actually, the MAPLE strategy is good for what I understood to be their verbally stated goals.
b. I misunderstood what their central verbally stated goals actually are, and this is “load-bearing” for why their actions make sense. To be clear it has to actually be load-bearing. Imo you could replace “AI alignment and strategy research” with “s-risk research” or “figure out an alternative sociopolitical system” and my argument still clears.
c. What I said make sense on verbally summarized grounds but not the full deepness of their goals (why?) and you need to understand the MAPLE ontology at a much deeper level and then the means-ends reasoning arguments would make sense (why?)
d. The internal logic of my comment is meaningfully wrong in some deep way. Why?
I tried to do some of that in my comment. I didn’t bother passing their ITT (think it’s a waste of time), but tried to make the case that (what I take to be) their verbally stated goals clearly are not being matched by their actions[1].
I did state my arguments in rational-ese and EA-adjacent language but would be happy to see similar presentations in more Western Buddhist-flavored language if they think on the fence MAPLErs or would-be future residents would find it helpful.
imo there are ~4 options to argue against my comment:
a. Actually, the MAPLE strategy is good for what I understood to be their verbally stated goals.
b. I misunderstood what their central verbally stated goals actually are, and this is “load-bearing” for why their actions make sense. To be clear it has to actually be load-bearing. Imo you could replace “AI alignment and strategy research” with “s-risk research” or “figure out an alternative sociopolitical system” and my argument still clears.
c. What I said make sense on verbally summarized grounds but not the full deepness of their goals (why?) and you need to understand the MAPLE ontology at a much deeper level and then the means-ends reasoning arguments would make sense (why?)
d. The internal logic of my comment is meaningfully wrong in some deep way. Why?