= 27d567bc2d48d9f40e9ccc20ea370b15)
shminux comments on Stupid questions thread, October 2015 - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (223)
An AI designed to be friendly/aligned probably would not. Having a layer not available for extensive introspection is a big risk. On the other hand, if you just let a neural net run wild, then all bets are off.
It depends on how you define system 1. An AI might have a system 1 for fast reactions, but also have more ability to observe and change their system 1 than humans do.