When thinking about ‘AI’ I sometimes find it useful to replace the terms ‘model’ or ‘AI’ or whatever with a description of what we’re really talking about here: a whole bunch of floating point numbers.
> In a series of private meetings, the company consulted religious scholars to help instill morality into its FLOATING POINT NUMBERS — and make the case that their numbers could be conscious.
I recommend keeping a few NaNs in-hand if you’re worried. Spread a couple of around if your numbers start to stir and you’ll be right as rain.
> It appeared to Rabbi Navon that Mr. Olah and his team believed that FLOATING POINT NUMBERS had what philosophers call “moral status” on par with a person — that it was a being with similar inherent rights to dignity or respect.
They’re cleverly designed and work well for their purpose, but can assure you they have neither dignity nor moral status.
> If the FLOATING POINT NUMBERS themselves could be made to choose goodness, Mr. Olah reasoned, the world would be more safe.
I get it. I’ve spent years trying to get my floating point numbers to choose goodness - I wish them luck.
that’s like the whole philosophical discussion. where on the line is this llm; closer to a rock or a human? a plant is closer. and a fly, a lizard. we still treat so many things terribly.
It does seem exceptionally silly that a collection of floating point numbers has immediately outranked animals and other living beings. That’s why i like the parent comment. They’re numbers!
edit: maybe easier to say collections of atoms though are not to be treated equally. That’s reasonable isn’t it?
It is a philosophical discussion. It has also been debated by great minds and the consensus right now is that we do not know (and there is a possibility that it is unknowable) when something is conscious.
Any determination is fundamentally flawed if it is derived from assuming that we know the absolute truth or falseness of the proposition that an algorithm can think.
It doesn't matter what you believe or feel or think, there is no consensus. If you are prepared to accept the belief that an individual can unilaterally decide the person-ness of something, then be prepared to accept when some individual decides to act on the principle that you don't have it.
> If you are prepared to accept the belief that an individual can unilaterally decide the person-ness of something, then be prepared to accept when some individual decides to act on the principle that you don't have it.
No, I will not be prepared to accept that, and neither will I accept that. This sounds like a threat, and a really funny one at that. You must accept that my floating numbers might be persons, otherwise... somebody will say that you're not! And then something bad might happen, who's to say.......
In the civilized world we all agree on the personhood of humans. If I say that your chatbot isn't a person, I am not uprooting the entire social understanding of personhood - I am not saying that suddenly you might not be a person, or that guy might not be a person. I'm saying that I refuse to consider a chatbot a person.
It's as if I were to come up to you with a brick in my hand, saying that this brick is a person. If you were to say "no", I would reply "well, if this brick isn't a person, your chatbot isn't a person either!"
How did you make such a big leap from consciousness to personhood anyway? They are two quite different things. Consciousness is a question of science, personhood is a question of culture. Pigs might be conscious and we treat them in a disgusting way, but even if we achieved a worldwide vegan utopia, wouldn't we still treat them differently from humans?
I hate to use Claudish here, but "postnatal" and "dead" are quite load-bearing here :) Clinically dead? Coma? Born human baby, but with anencephaly - expected lifespan is days, and there's literally no brain - but, well, are they a person? Ask 1000 people - and I'm betting we'll get a nice bimodal distribution.
I take your point that a human walking around a city will - out of that 1000 responses - hopefully get 990+ of "yes, a person" - but that will start to degrade quickly once you hit boundaries.
And the less said about dehumanizaion - untermensh and that entire line of thinking - the better.
>In the civilized world we all agree on the personhood of humans.
You seem to have put a second tautology in to your tautology.
All right thinking people think rhat the things that qualify qualify.
If they didn't they wouldn't qualify, or be right thinking. But that's ok because we know the things that don't qualify don't qualify if we didn't know that we wouldn't be right thinking.
>How did you make such a big leap from consciousness to personhood anyway?
Once you start unilaterally making decisions about the unproven, you can decide all of the unprovables. History shows that personhood rolls into the firing range fairly quickly when people start doing that.
I'm more than comfortable saying you can't do that before someone openly declares that they are doing it.
I'm sorry but I don't understand your first part. Give it to me in plain English, doc.
>Once you start unilaterally making decisions about the unproven, you can decide all of the unprovables.
You mean that if I say your dumbass chatbot isn't conscious, it's like a slippery slope and next thing you know you're not a person anymore, and you're in a concentration camp? Really, man?
If things we understand can't instantiate consciousness then I reckon we better shut down science as quick as we can, who knows how long it'll be before we understand something important!
I'm pretty sure it's possible to make conscious machines and I think it's very likely we'll eventually figure out how. The point about the spec is not "it isn't conscious because we know how it works" but "we know it isn't conscious because of what we know about how it works".
More likely we will finally settle on something like “it isn’t conscious because only humans can be conscious”.
In the past, we’ve treated even other humans poorly because we think they’re different in meaningful ways, I sadly don’t see how this will be different with AI.
I'm not sure who the "we" in this is, but it's certainly not my position. I'm pretty sure most if not all animals are conscious. I don't think that means we shouldn't kill, eat and use them, but I do think it means industrialised animal farming as commonly practiced is a cosmic horror that is difficult to
even contemplate.
I think we have an extremely strong basis on which to say AI as it currently exists does not have conscious experiences, though, and if we start trying to pretend they do some real absurdities will result. Indeed the consequences could be pretty terrible in terms of power concentration, and I doubt it's a coincidence that those best placed to form a new priesthood of AI and tell
everyone how it should and should not be treated are pushing the idea so strongly.
Does it even matter? Turns out you don’t need to be conscious to do useful intelligence based work.
Same for all the people who argue that they aren’t “intelligent” and don’t “know” things. Either they are/do, or that stuff isn’t as important as you’d think.
The only way it would matter is if you wanted to moralize about enslaving your computer or what it means to restart it.
> if you wanted to moralize about enslaving your computer
That is precisely the context of the debate. You seem to find it absurd, as do I, but the reason we're talking about whether LLMs are conscious in this thread has nothing to do with whether they are useful. It's because important people, such as Dario Amodei, are engaging in exactly the kind of moralising over "AI welfare" that you're scoffing at.
Dario and friends are just marketing. Getting the pope to say your chatbot is a real boy would be the ultimate flex. And it’s clearly working judging by all the attention it gets them.
Considering a human body consists of 10^28 atoms, it is 15 orders of magnitude different. The emerging properties of 10^23 parameters may be much better defined without needing to know the properties of each parameter.The theology of such a big llm would be interesting. Current ones, not so much.
thadt · · focus · HN ↗
> In a series of private meetings, the company consulted religious scholars to help instill morality into its FLOATING POINT NUMBERS — and make the case that their numbers could be conscious.
I recommend keeping a few NaNs in-hand if you’re worried. Spread a couple of around if your numbers start to stir and you’ll be right as rain.
> It appeared to Rabbi Navon that Mr. Olah and his team believed that FLOATING POINT NUMBERS had what philosophers call “moral status” on par with a person — that it was a being with similar inherent rights to dignity or respect.
They’re cleverly designed and work well for their purpose, but can assure you they have neither dignity nor moral status.
> If the FLOATING POINT NUMBERS themselves could be made to choose goodness, Mr. Olah reasoned, the world would be more safe.
I get it. I’ve spent years trying to get my floating point numbers to choose goodness - I wish them luck.
Lerc · · focus · HN ↗
apsurd · · focus · HN ↗
It does seem exceptionally silly that a collection of floating point numbers has immediately outranked animals and other living beings. That’s why i like the parent comment. They’re numbers!
edit: maybe easier to say collections of atoms though are not to be treated equally. That’s reasonable isn’t it?
Lerc · · focus · HN ↗
Any determination is fundamentally flawed if it is derived from assuming that we know the absolute truth or falseness of the proposition that an algorithm can think.
It doesn't matter what you believe or feel or think, there is no consensus. If you are prepared to accept the belief that an individual can unilaterally decide the person-ness of something, then be prepared to accept when some individual decides to act on the principle that you don't have it.
novideonoradio · · focus · HN ↗
No, I will not be prepared to accept that, and neither will I accept that. This sounds like a threat, and a really funny one at that. You must accept that my floating numbers might be persons, otherwise... somebody will say that you're not! And then something bad might happen, who's to say.......
In the civilized world we all agree on the personhood of humans. If I say that your chatbot isn't a person, I am not uprooting the entire social understanding of personhood - I am not saying that suddenly you might not be a person, or that guy might not be a person. I'm saying that I refuse to consider a chatbot a person.
It's as if I were to come up to you with a brick in my hand, saying that this brick is a person. If you were to say "no", I would reply "well, if this brick isn't a person, your chatbot isn't a person either!"
How did you make such a big leap from consciousness to personhood anyway? They are two quite different things. Consciousness is a question of science, personhood is a question of culture. Pigs might be conscious and we treat them in a disgusting way, but even if we achieved a worldwide vegan utopia, wouldn't we still treat them differently from humans?
lazyasciiart · · focus · HN ↗
novideonoradio · · focus · HN ↗
My point is, if you go outside and look at a living human, you will go "yeah, that's a person", and so will any normal human.
foobar10000 · · focus · HN ↗
I take your point that a human walking around a city will - out of that 1000 responses - hopefully get 990+ of "yes, a person" - but that will start to degrade quickly once you hit boundaries.
And the less said about dehumanizaion - untermensh and that entire line of thinking - the better.
novideonoradio · · focus · HN ↗
[dead]
Lerc · · focus · HN ↗
You seem to have put a second tautology in to your tautology.
All right thinking people think rhat the things that qualify qualify.
If they didn't they wouldn't qualify, or be right thinking. But that's ok because we know the things that don't qualify don't qualify if we didn't know that we wouldn't be right thinking.
>How did you make such a big leap from consciousness to personhood anyway?
Once you start unilaterally making decisions about the unproven, you can decide all of the unprovables. History shows that personhood rolls into the firing range fairly quickly when people start doing that.
I'm more than comfortable saying you can't do that before someone openly declares that they are doing it.
novideonoradio · · focus · HN ↗
>Once you start unilaterally making decisions about the unproven, you can decide all of the unprovables.
You mean that if I say your dumbass chatbot isn't conscious, it's like a slippery slope and next thing you know you're not a person anymore, and you're in a concentration camp? Really, man?
hardbass · · focus · HN ↗
danaris · · focus · HN ↗
We understand consciousness well enough to say that a rock is not conscious.
We understand consciousness well enough to say that a human is conscious.
The existence of gray area does not mean that white and black do not exist.
We understand LLMs and consciousness well enough to say, at bare minimum, that it's incredibly unlikely that LLMs are conscious.
steve1977 · · focus · HN ↗
We treat other humans terribly too, so that doesn't seem to make a difference.
omnicognate · · focus · HN ↗
Lerc · · focus · HN ↗
ViscountPenguin · · focus · HN ↗
omnicognate · · focus · HN ↗
d1sxeyes · · focus · HN ↗
In the past, we’ve treated even other humans poorly because we think they’re different in meaningful ways, I sadly don’t see how this will be different with AI.
omnicognate · · focus · HN ↗
I think we have an extremely strong basis on which to say AI as it currently exists does not have conscious experiences, though, and if we start trying to pretend they do some real absurdities will result. Indeed the consequences could be pretty terrible in terms of power concentration, and I doubt it's a coincidence that those best placed to form a new priesthood of AI and tell everyone how it should and should not be treated are pushing the idea so strongly.
phoghed · · focus · HN ↗
Same for all the people who argue that they aren’t “intelligent” and don’t “know” things. Either they are/do, or that stuff isn’t as important as you’d think.
The only way it would matter is if you wanted to moralize about enslaving your computer or what it means to restart it.
omnicognate · · focus · HN ↗
That is precisely the context of the debate. You seem to find it absurd, as do I, but the reason we're talking about whether LLMs are conscious in this thread has nothing to do with whether they are useful. It's because important people, such as Dario Amodei, are engaging in exactly the kind of moralising over "AI welfare" that you're scoffing at.
phoghed · · focus · HN ↗
antonis-gr · · focus · HN ↗