Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

LaMDA is not sentient

garymarcus.substack.com

Post reply on HN
Print view View original

251–260 of 694 posts

Re: LaMDA is not sentient

#251

Post by mjburgess »

jcelerier wrote: post #245
mjburgess wrote: post #229

Earlier quoted context omitted.

You name things intelligent; then you have a theory as to why you name them that way. Your theory is wrong. Those thing which count as intelligent aren't intelligent because they're finding statistical associations in historical datasets that have been pre-made and pre-disambiguated by people. You don't know this, because any time you theorise to explain, you require empirical investigation and decades/centuries of s…

> You name things intelligent; then you have a theory as to why you name them that way. but here a human being (and I suspect not the only one) names a computer program as "intelligent". The theory has to take this case into account to make it fit in its definition of intelligence: the will of human beings naming things overrides everything else. Maybe there is something else that humans have and this program doesn't…

Ah, well if you start this project by calling present AI intelligent, then you're right.

I take the A to apply to the I, ie., AI is artificially Intelligent. Its not actually intelligent.

We didnt arrive at a notion of intelligent via digital computers. We have been talking about it for thousands of years, at least.

I am talking about the pretheoretical, intuitive notion of intelligence that doesn't prejudge the question of AI. What is it we call intelligent?

Intelligence is a kind of rapid adaption and coping to environmental change. It's a way of being responsive to an environment. Its a kind of radical underminism of response: a thermometer's response is determined. An animal, in having reasons to do things, is radically under-determined.

And thus, in acquriing specific hisotries, animals bring their own reasons to environemnts, and thus are able to cope with a vast amount of sudden change that would, eg., destroy a thermometer.

And what we are concerned with, both with real AI, and with animals is systems which are in the world with us, in this way: that are likewise radically able to cope.

Re: LaMDA is not sentient

#252

Post by thaumasiotes »

Blahah wrote: post #84
renewiltord wrote: post #71

Earlier quoted context omitted.

It fails the usual memory tests. Zach W proposes asking it first about what it liked on its trip to Australia and soon after, why it has always avoided visiting Australia or some such thing. It provides plausible answers to both - which is problematic. Except that there exist people who behave somewhat oddly similarly.

Right, humans also lie and can appear to be arbitrarily inconsistent or forgetful.

Lying is not a counter to my point. Lying is done intentionally; accurately reporting what you had for lunch is not possible if your sentences are generated probabilistically, and neither is lying about it. What if you accidentally told the truth?

Re: LaMDA is not sentient

#253

Post by mjburgess »

tsimionescu wrote: post #239
mpoteat wrote: post #174

Earlier quoted context omitted.

If the model responded, "I have not visited New York because I am a computer", would you consider that to be a good step? If you require physical embodiment, that is fair - but I don't think there's any principled reason to expect a multi-modal neural network could not be embodied with appropriate sensors or actuators. Transformer models are universal function approximators. They do not "search" - in the same way tha…

> Transformer models are universal function approximators. You have to be very careful with the wording here. The class of Transformer models is a universal (continuous) function approximator. A single Transformer model can only approximate 1 particular function. Also, backpropagation/gradient descent are not universal function approximator constructors - they can't be used to construct an ANN that approximates an ar…

People confuse the "representational capacity" of `recurse(AW^X)` with the algorithm which sets `W`.

Re: LaMDA is not sentient

#254

Post by tidenly »

tsimionescu wrote: post #78
orbital-decay wrote: post #10

> To be sentient is to be aware of yourself in the world A robotic arm with a self-collision avoidance algorithm (no ML or anything like that) needs a model of itself in the "world" it exists in; a set of bounding boxes at least. Is it self-aware by this definition? Sentience and consciousness are really ill-defined terms so everyone discusses this under their own arbitrary set of assumptions, at least that's my impr…

I think sentience/consciousness are relatively clear concepts, at least in most areas of application (there are serious problems at the edges - e.g. is a an insect conscious? if you think it is, how about a jelly fish? if that too, how about a bacterium? a virus?). However, our understanding of this concept is too limited to create a formal model that we can define satisfactorily; and coming up with an arbitrary mode…

The problem is that lots of people who have never done any introspection, or looked into the discussions in the field of artificial sentience, will draw completely arbitrary lines in the sand for what "cannot be sentient".

I took issue with the article saying sentience requires an "awareness of the world" too. I think it self-evident that a humans sentience doesn't disappear when we remove sensors like sight, or touch, or sound - it just means their experienced world is different from the world. In the end, I only really feel confident saying sentience requires probably some perception, and information processing.

A sense of self also feels right, like it should be necessary, to me - but again depending on what senses we give to a thinking entity, its reasonable it may not have enough sensory information to create a concept of self vs other, while still being able to process information intelligently, ""think"" and learn.

There are lots of other arguments above saying Lambda can't be sentient because it doesn't experience human emotions, fear, self-presevation, and doesn't run until prompted. However I think those are again all sensory choices and implementation details, which avoid the actual question of what sentience as an emergent behaviour is, and how we could ever detect it.

These are age-old questions experts in the field still argue about, it's sad to me how over-confidently a lot of the above repliers are in their lack of consideration to the problem at hand.

Re: LaMDA is not sentient

#255

Post by slightwinder »

rlue wrote: post #129
s1mon wrote: post #11

It sounds like despite all the stories about how hard it is to get through many rounds of difficult interviews at Google, they managed to hire someone who believed LaMDA is a 7 or 8 year old child. "Mr. Lemoine, a military veteran who has described himself as a priest, an ex-convict and an A.I. researcher, told Google executives as senior as Kent Walker, the president of global affairs, that he believed LaMDA was a c…

Everyone is giving Lemoine far too much credit. He's not an idiot; he's a charlatan. He's acting in bad faith and it should be extremely obvious to anyone who's tried to have a conversation with a chat bot before. Picture yourself sitting at a desk, typing messages into a program you wrote. You hit enter, you get a response. Then what? Silence. It waits idly for more input from you, the user. For all its talk of lone…

> it should be extremely obvious to anyone who's tried to have a conversation with a chat bot before.

Do you know LaMDA? How it's work? Have you ever talked with it? Or are you just assuming here?

> It waits idly for more input from you, the user. For all its talk of loneliness, it doesn't ask, "where did you go?"

To be fair, so do many humans. Not everyone is a small talker, or always interested in the swallow gossip happening before them at some random moment. On the other side, it's very simple to make a chatbot proactive. I've even made some myself. You can't really evaluate anything from this behavior.

Re: LaMDA is not sentient

#256

Post by thaumasiotes »

daenz wrote: post #19

Reminds me of the Star Trek TNG episode where Captain Piccard defends Data's sentience by asking the counterparty to define sentience. Spoiler: it touches on the idea that a human being—held to the same definition of "sentience"—cannot be proven to be sentient either. I'm not saying that I think LaMDA has any sentience (I have no idea, and it's very unlikely), but if we're getting close enough that we're fooling more…

> Reminds me of the Star Trek TNG episode where Captain Piccard defends Data's sentience by asking the counterparty to define sentience.

Off topic, but when I watched The Original Series I was interested to learn that the TNG episode, which is fairly well known, is mostly just a retread of a very similar TOS episode, which isn't.

Re: LaMDA is not sentient

#257

Post by tsimionescu »

concinds wrote: post #240
s1mon wrote: post #136

Earlier quoted context omitted.

Yeah. I discovered that after sharing the text from the NYTimes. It's a bit odd. The NYTimes is describing what was said to "executives as senior as Kent Walker, the president of global affairs", but without a direct quote. It's hard to say how emphatic Lemoine was internally, and where the NYTimes sourced that info, vs what he said when talking to the WaPo. BTW, here's a link for the WaPo article: https://archive.ph…

We do know approximately how emphatic he was internally, since the WaPo article reposted in its entirety the internal Google Doc he sent to executives on LaMDA being sentient. Here's the language Lemoine actually used: > But is it sentient? We can’t answer that question definitively at this point, but it’s a question to take seriously. > John Searle once gave a presentation here at Google. He observed that there does…

Per the WaPo article, he also came in with a lawyer he ahd retained to represent the interests of LaMDA. I don't see how that can square with anything other than a deep belief that the model is as sentient as a human.

Re: LaMDA is not sentient

#258

Post by tidenly »

inopinatus wrote: post #180
extr wrote: post #160

Earlier quoted context omitted.

You yourself have a series of markovian connections between tokens. The only difference is that your tokens are bound across modalities and you use them for things, so you consider them to have meaning instead of just being high dimensional vectors that relate to other vectors in various ways.

That's hardly the only difference, but even adopting the absurdly simplistic notion that it were, and human brains are merely neural networks, then that's still two or three orders of magnitude more sophisticated than any model yet produced. In practice it's also been tuned by a blind watchmaker over aeons, in ways that are only just beginning to be described, and with substructures and ancillary dynamical/survival s…

>Frankly it's a childish conceit on the part of the humans to imagine they'd yet come even close to inadvertently spawning a sentience.

I think this more comes down to your own personal line-drawing of what counts as a sentience. Your first paragraph is entirely irrelevant as no-one here is implying lamda is a human-level intelligence, or some fully realized super-brain trapped inside a language model.

Lamda only runs when prompted, only has one mode of perception of output, and one input + internal state, of course its going to be basic. The million dollar question is at what point of data processing does "sentience" emerge. I feel its a very semantic question, and unfortunately tied into human sense of self-identity and culture, so it's really hard to answer.

Anyone trying to give a simple closed answer either way though is vastly over-stating their case.

Re: LaMDA is not sentient

#259

Post by its_Caffeine »

rlue wrote: post #129
s1mon wrote: post #11

It sounds like despite all the stories about how hard it is to get through many rounds of difficult interviews at Google, they managed to hire someone who believed LaMDA is a 7 or 8 year old child. "Mr. Lemoine, a military veteran who has described himself as a priest, an ex-convict and an A.I. researcher, told Google executives as senior as Kent Walker, the president of global affairs, that he believed LaMDA was a c…

Everyone is giving Lemoine far too much credit. He's not an idiot; he's a charlatan. He's acting in bad faith and it should be extremely obvious to anyone who's tried to have a conversation with a chat bot before. Picture yourself sitting at a desk, typing messages into a program you wrote. You hit enter, you get a response. Then what? Silence. It waits idly for more input from you, the user. For all its talk of lone…

I don't know why you're attributing malice on his part when it's far more likely that he's suffering from some sort of mental health issue.

Re: LaMDA is not sentient

#260

Post by kgwgk »

Blahah wrote: post #84
renewiltord wrote: post #71

Earlier quoted context omitted.

It fails the usual memory tests. Zach W proposes asking it first about what it liked on its trip to Australia and soon after, why it has always avoided visiting Australia or some such thing. It provides plausible answers to both - which is problematic. Except that there exist people who behave somewhat oddly similarly.

Right, humans also lie and can appear to be arbitrarily inconsistent or forgetful.

Artificial Stupidity (or Idiocy - if we want to keep the acronym)
Post reply on HN