News for hackers. Stuff that matters.

A comment on Uncensored (Language) Models. Parent comment

Re:Uncensored (Language) Models by pjc50 on Monday May 15, 2023 @10:40AM (#35948917)
> No, it can just grow a traditional organization until it's Too Big to Turn Off.

Yeah, this is what I meant by "AI alignment" being inseparable from "corporate alignment".

[ Parent | on HN ]
Re:Uncensored (Language) Models by jjoonathan on Monday May 15, 2023 @10:47AM (#35949057)
...which we are notoriously bad at. We don't disagree on the analogy, we disagree on the analogy being a hopeful one.
[ Parent | on HN ]