niplav @niplav@schelling.pt

The Amish should be able to continue doing whatever they're doing

Nyfb vs nag yvirf ner pheeragyl arg artngvir gurz tbvat rkgvapg jbhyq or n tbbq guvat.

my intuition: V guvax nagf jbhyq or rvgure rkgvapg be bayl nyvir va mbbf &p. Gur rpbabzl gura jbhyq or ~43 beqref bs zntavghqr ovttre, naq V guvax n ybg bs gung jbhyq'ir fgvyy unccrarq ba rnegu. Uhznaf qba'g pner rabhtu nobhg nagf gb xrrc gurz nyvir gung zhpu (rfcrpvnyyl fvapr jr'q rkcrpg uhzna qrfpraqnagf gung sne vagb gur shgher pner znvayl nobhg ercebqhpgvir svgarff qverpgyl).

if economic growth continued at 1% a year, for 10k years, would that be good or bad from the position of ants in 10k years?

law of one player: nothing happens unless you (yes, you specifically) make it happen

common theme rn in alignment research is how value evolve, e.g. concept/value extrapolation, shard theory

if you're german then the term "gaslighting" is slightly queasy

can you money-pump the agents simulated by large language models?

give standard vNM coherence violating scenarios to smarter language models

at higher levels of intelligence, maintaining coherence becomes more difficult since your action space widens and having high coherence might be NP-hard. so the better metric is coherence divided by size of action space

your mom is the optimization target

love languages as multi-armed bandits

onrushing tide, a mere
fifty miles away, and just
our puny channels

should i start a podcast

https://en.wikipedia.org/wiki/Returning_soldier_effect