Comments by "xybersurfer" (@xybersurfer) on "Is AI a threat to human civilization? | Manolis Kellis and Lex Fridman" video.

@momentary_ that's one of the problems with these Large Language Models. it's a black box that they are still researching how it works. there is essentially no code. aside from that "well being" is rather vague and does not easily translate directly to computer code. you might be able to instruct it about "well being", but that does not mean that it will listen the way you would want
1