4 ms·
> I find it odd that discussions of alignment don't mention 'legality' Because other wishy-washy stuff doesn't involve prison. Not that our new oligarchs with
by noisy_boy 2mo ago
> I find it odd that discussions of alignment don't mention 'legality'
Because other wishy-washy stuff doesn't involve prison. Not that our new oligarchs with the politicians in their pockets have any real risk of it, but why take chances. Much safer to doodle about alignment and such abstractions in safer and softer contexts.
- anentropic 2mo agoBut that was kind of my point Maybe it'd be easier to train the models on key parts of the legal code and give it a hard aversion to breaking the law - rather than training on vague value judgements and then hope the model doesn't break the law
- noisy_boy 2mo agoI got that. I'm saying that it is deliberate. Wiggle room et all.
- anentropic 2mo agoHow does not explicitly trying to train the model specifically not to break the law give them any wiggle room if the model then goes and breaks the law?
- noisy_boy 2mo agoDeniability on intention. Models can hallucinate, all bets are off. Sure, they can make it stricter but why do that and subject themselves to harder scrutiny based on the training criteria?