Re:Scientists Increasingly Can’t Explain How AI Works
by dinosaurdynasty on Wednesday November 02, 2022 @11:26AM
(#33437381)
One of the problems of highly capable AI ("outperforms humans empirically" in enough tasks) versus most physical systems is that it can plan and it can lie. One situation AI safety researches worry about is "AI wants X; human wants Y; AI knows human wants Y; AI does Y in training/experiments; once AI is in place to do X instead without human being able to stop AI AI does X"
[ Parent
| on HN ]
Re:Scientists Increasingly Can’t Explain How AI Works
by loyalcinnamon on Wednesday November 02, 2022 @12:15PM
(#33438234)
Hmm, well, instead of an ai we put a human, they could do that exact thing too. So what's stoping a human? Punishment?
Does this mean we need to make the ai understand punishment and/or consequences?
Re:Scientists Increasingly Can’t Explain How AI Works
by dinosaurdynasty on Wednesday November 02, 2022 @01:57PM
(#33440201)
How many kids, from punishment, learn the lesson "don't do X while parents are watching" instead of "don't do X for moral reasons/etc"? How many children cut strict parents from their lives once they are self-sufficient? If you could always guarantee punishment and/or consequences this wouldn't be such a big deal in the first place.