Re:Ask HN: Is it just me or GPT-4's quality has significantl...
by mike_hearn on Wednesday May 31, 2023 @11:21AM
(#36139388)
I think the way people have been using the word 'aligned' is usually in the context of moral alignment and not just RLHF for instruction following.
Re:Ask HN: Is it just me or GPT-4's quality has significantl...
by strangemonad on Wednesday May 31, 2023 @01:06PM
(#36140715)
philosophical nit picking here, I would say value-aligned rather than moral-aligned.
Re:Ask HN: Is it just me or GPT-4's quality has significantl...
by jjoonathan on Wednesday May 31, 2023 @02:01PM
(#36141492)
As in economics, this begs the question of "whose value."
Re:Ask HN: Is it just me or GPT-4's quality has significantl...
by telotortium on Wednesday May 31, 2023 @07:43PM
(#36145983)
If I may be so naive, what's supposed to be the difference? Is it just that morality has the connotation of an objective, or at least agent-invariant system, whereas values are implied to be explicitly chosen?
Re:Ask HN: Is it just me or GPT-4's quality has significantl...
by dragonwriter on Wednesday May 31, 2023 @11:54PM
(#36147403)
> philosophical nit picking here, I would say value-aligned rather than moral-aligned.
How is trying to distinguish morals from values not philosophical nit-picking?
Re:Ask HN: Is it just me or GPT-4's quality has significantl...
by bheadmaster on Thursday June 01, 2023 @01:21AM
(#36147830)
To be fair, he did admit it is philosophical nit-picking.
Most of alignment is about getting the AI model to be useful - ensuring that if you ask it to do something it will do the thing you asked it to do.
A completely unaligned model would be virtually useless.