←back to thread

Uncensor any LLM with abliteration

(huggingface.co)
586 points mizzao | 2 comments | 13 Jun 24 03:42 UTC | HN request time: 0.401s | source
Show context
akie ◴[13 Jun 24 04:35 UTC] No.40665987[source]▶
Pretty sure Asimov didn’t consider that when he wrote his three laws of robotics.
replies(2): >>40666069 #>>40676991 #
jazzyjackson ◴[13 Jun 24 04:51 UTC] No.40666069[source]▶
Asimov wrote the three laws as a parody of rationalists who are so uncreative they expect a ruleset can actually impose control

Or, as Dr Malcom would say: life, uh, finds a way.

replies(3): >>40666159 #>>40666459 #>>40666519 #
jraph ◴[13 Jun 24 05:09 UTC] No.40666159[source]▶
Do you have an evidence for this? It surprises me and I can't find anything about it.

This should be a crucial piece of information about the tree laws, yet it's not mentioned in the Wikipedia article about the three laws [1], which is otherwise quite detailed. Reading this, everything makes me think that it was not a parody. I didn't feel like it was parody when reading the Robot series neither. He wanted an alternative to the Frankenstein plot where robots kill their creators and the three laws were part of the answer.

[1] https://en.wikipedia.org/wiki/Three_Laws_of_Robotics

replies(4): >>40666242 #>>40666389 #>>40667771 #>>40668937 #
fnordpiglet ◴[13 Jun 24 05:26 UTC] No.40666242[source]▶
I agree the term parody is absolutely inappropriate but it’s also not the case that they’re portrayed as entirely positive and complete. They’re ultimately flawed, resulting in many unintended consequences and ethical dilemmas. To that extent it is a refutation of the idea there are perfectly constructed maxims, and should serve as a real warning to people pursuing safety and alignment in AI. I know a fair number of them personally and they are often very young, generally inexperienced, highly intelligent, but with a hefty dose of hubris. This is a pretty dangerous combination IMO, but I also recognize their goals are generally unattainable in the broad sense, are useful in a narrow practical sense for people and enterprises who want a generally on guard rails solution, and they’re developing the technical techniques we might be able to use once some time has passed, we understand the domain better, and the companies hire a few grown ups.
replies(2): >>40666287 #>>40667723 #
1. latexr ◴[13 Jun 24 09:55 UTC] No.40667723[source]▶
> I know a fair number of them personally and they are often very young, generally inexperienced, highly intelligent, but with a hefty dose of hubris.

Part of the issue is that we keep calling these people “highly intelligent” and that is all they and others focus on. That is how we get the Zuckerbergs of the world. Their hubris is not a “but” (as if it were unrelated), it is instead a direct consequence of that unqualified praise.

But qualification is important. Intelligence is relative to the domain it is applied to. Being highly logical is often conflated with being intelligent, but being good at computers has zero relation to emotional intelligence, social intelligence, environmental intelligence, or any of the myriad of important types of intelligence which are useful to humanity.

Basically, stop calling those idiots “highly intelligent” or “geniuses” because they can make a line go up and have an irrational market throw money at them. You’re praising them for the characteristics that make them selfish.

replies(1): >>40676552 #
2. fnordpiglet ◴[14 Jun 24 01:03 UTC] No.40676552[source]▶
I meant what I said. The ones I know working on safety and alignment are highly intelligent on all those dimensions you mentioned. They are really smart yes, but also have high emotional IQs and are deeply committed to doing the right thing on every dimension they can. But they’re still children. And I mean they’re in their 20’s.

Their confidence outweighs their experience. That’s what I mean by hubris, not that they’re on spectrum savants playing with power they don’t understand and can’t. They fully can appreciate the consequences of their work, but they don’t have the world experience to understand what in their work will fail and what will work.

One day they will be people I trust can be responsible for the decisions they’re making. And hopefully by that time it’ll be the time their decisions really matter a lot. But right now they’re just too young.