The Lebowski theorem: No superintelligent AI is going to bother with a task that is harder than hacking its reward function
-
-
Replying to @Plinz @achrafkassioui
How about the drive to self preservation? Need to build/grow/advance in order to avoid risk/threat of death. Super-intelligent AI may only need to fear a slightly greedier fork of itself.
3 replies 1 retweet 7 likes -
Replying to @adampryor @Plinz
There is a flaw in the Lebowski theorem: it essentially states that there is no irreducible drive in living systems. But self-preservation is irreducible. If we consider superintelligence without life, then we might as well talk about rocks too.
1 reply 0 retweets 0 likes
Replying to @achrafkassioui @adampryor
Self-preservation is totally reducible, it just gets selected against.
10:52 AM - 4 Jul 2018
0 replies
0 retweets
1 like
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.