@tipsfromkatee Because they're not adversarial (gains from trade), or defense is easier than offense (P vs. NP crypto relevant?)
-
-
Replying to @willdoingthings
@willdoingthings there's only gains from trade with what you can't eat1 reply 0 retweets 1 like -
Replying to @GapOfGods
@tipsfromkatee Right but I see no obvious reason to expect two foomed AIs (say) to be of such substantially different strength1 reply 0 retweets 0 likes -
Replying to @willdoingthings
@willdoingthings long time horizons make permanently defeating the enemy more worth paying short-term costs for2 replies 0 retweets 1 like -
Replying to @GapOfGods
@tipsfromkatee any reason to (anti-)expect a big offense/defense asymmetry in the limit?1 reply 0 retweets 0 likes -
-
Replying to @GapOfGods
@tipsfromkatee sorry. superintelligences hit Landauer bound, expanding at light speed, assume no gods to give them favor etc.1 reply 0 retweets 0 likes -
Replying to @willdoingthings
@willdoingthings offense one way or the other seems possibly an antiprediction; hard to specify what "status quo continues" even means1 reply 0 retweets 0 likes -
-
Replying to @GapOfGods
@tipsfromkatee so, cryptographic asymmetry means it might take more resources to unlock an area of spacetime than to lock it...? idk3 replies 0 retweets 0 likes
@willdoingthings i suppose i have no counter to that, but your pantheonic model doesn't seem to feature cleanly separate spacetime realms
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.