I operate by Crocker's rules[1].
[1]: https://www.lesswrong.com/tag/crockers-rules
ne konfuzu “ilo” kaj “ulo”—uloj ne ŝatas ĝin.
Oh boy "be willing to airstrike datacenters" oh no ah stumbling I think
(In case this sounds like a promise: It's also not a very good joke)
Should I post a very tasteless joke
best yud
interesting twitter FOMO
lesswrong request rate limit exceeded
heh
I've never gotten much out of a book review & don't understand why people read or write them.
shifting from direct to indirect realism is important for value extrapolation
but can you trick the plan evaluation evaluation procedure?
CIRLHF
STATISTICAL REGULARITIES IN PEOPLESPACE
Whether or not AI systems undergo fast takeoff is a fact of computer science, not a strategy
THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT A HIGHLY ESTEEMED DEED, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO BEING A PLACE OF HONOR, FITNESS FOR COMMEMORATION, AND ANYTHING OF VALUE
Shrimps
Well, slightly, since game theory is so rarely applicable
This actually cruxes my position on explotation
For a normal-form game G and a player i, can removing actions from player i yield a better Nash equilibrium *for i*?
Has this been investigated?
Have you ever completely read a thing I've written on my site?
@pernia I will post until the day I die
a Schelling point for those who seek one