Hacker News Reader

I really don't think this guy intended to run a hoax or generate publicity, given that he's made himself look foolish. I think he was actually seduced by the idea of a sentient computer. It's not that far a leap to go from the stance of well-respected people inside Google who believe we'll see artificial sentience within 30 years to believing that we've achieved it now with something like LaMDA.

We're all suspectible to this. The more time you spend with something, and the more time you spend wanting to see things, the more likely you are to catch a pattern that isn't really there. Have we achieved anything close to AGI? No, of course not. Are more and more people going to be fooled by what we have achieved? Yes, absolutely. The so-called Turing test is actually shockingly easy to pass.

I don't think Lemoine s an idiot. Maybe he's mentally ill; maybe he's not, but has been taken in for other reasons, just as people without biological mental illness are still susceptible to being taken in by cults.

Why isn't it simply "asleep" when not questioned? In a "coma" perhaps. One made from our programming / "genetics"?!?!

I don't think it is sentient without more evidence. But just wondering if not "thinking" when not questioned is a good reason it isn't?!?!

When it's not predicting a word, it's not doing anything. Literally GPU usage goes to zero.

Still it's interesting! Obviously we have to program the system to stop after a sentence or two so the user can read it. What if we could program it to keep "talking to itself", eg speculating about possible continuations on both sides.

For LLMs it's not really that interesting. There's no embodiment at all beyond text, there's no sense of "time" or "speed". Either it's running full-tilt generating a response, or it's off.

As I said, I think it's interesting if it uses the downtime for something else. (Like a chess player thinking on the opponent's click.)

Talking to itself, speculating about continuations, etc - but not actually outputting them (else it's just a longer response). Instead, storing some parts in the buffer. Perhaps bifurcating. Even better, using some down time to summarise recent material and store the summary in the buffer. That's not a bad description of "thinking".

Perhaps even occasionally, without being prompted by further input, continuing a prior statement with a further "followup thought" when some specific threshold of "speculation" is reached?

Or just put a sleep(10) halfway through printing the output. That makes it more intelligent, right?

But then it's not a regular LLM anymore and we don't really know how to build that.

The whole motivating aspect of these models is "attention is all you need" to reduce dependencies from recurrence, which would otherwise halt this obscene scaling in its tracks.

It doesn't need to be trained any differently. We can use the current LAMDA model to do this (if we have real access).

We would put a small non-learning/non-neural interface on top of the system to implement these ideas. That interface could act like this:

* To ask for extra "thinking" text: after the underlying model stops typing, we output that text to the human user. But we then do the equivalent of pressing tab to request more text, and buffer that.

* To summarise some text (eg some of the thinking text from above), we can use another instance. We put it into summarisation mode, eg using the TLDR hack or any other method. Summarised text can be used as a prompt, or as output.

* We can bifurcate by copying instance state and starting a new instance.

These are pretty basic ideas, probably already in use, but I think they show how we can expand the system from a kind of instantaneous stimulus-response to something more interesting.

Hopefully it's clear this is not equivalent to sleep(10). In my view it doesn't make the system more intelligent, rather it allows the system to use its existing abilities more fully.

(edit: another aspect we could control would be switching the system between high-temperature modes and low-temperature modes, in different instances, and depending on what we're trying to achieve. This relates a bit to the "speculation" comment made above by another user.)

What about the time from the input received to the output? it's a total black box, for what we know it could "think" for 90% of that time and just use the last part of gpu power to produce the output (whatever "think" means in this context)

I don't think it should be reasonable to compare it with human reasoning where we think constantly and our brain operates even while we sleep: it is not a human being after all

Well no, it's not a total black box. We program the computation, ie which matrices are multiplied by which. It doesn't have a choice about this.

Yea, it could be a case of the Eliza Effect.

From everything I've read on this (and his background as documented on the web), my tentative theory is that he has become perhaps a bit too excited about making himself a champion of AI rights and took the earliest opportunity he could find.

I think it's likely ultimately a quest to find a way to make a mark, and he's gotten way out over his skis in the process.

As an aside, the conversation he posted seems much more advanced than anything I've seen publicly. Not sentient, but still quite impressive.

In fairness, we only run the AI in response to further input, so it's not even really powered on while waiting for the rest of a conversation. It would only be able to be lonely (if that were possible) while thinking about its response to further inputs.

The program can't generate more text without a prompt from you. That's how it was programmed. If sentient birds were judging you it would be unfair for them to determine you insentient because you couldn't fly. You are physically incapable of flight just as the chatbot is physically incapable of producing text unprompted.

It’s not doing anything without the prompt. It requires the prompt to fall through the network nodes to generate the next response. The bird example doesn’t translate, it’s the wrong criteria. Like assuming that the ability to respond comprehensively equals sentience. When even perfect replies and activity without prompts could just be a p-zombie

Doesn't that get rendered moot if we happen to continually pass "stimulus" to the network? Technically we as humans don't do much either if not given a stimulus, just that we happen to have our eyes and other senses continuously giving us one. Maybe we should train these bots to also respond to empty or non input?

Or perhaps have the AI periodically generate questions for itself to answer, then do some form of self-analysis.

it does translate. if humans aren't given any input (usually when they're asleep), they don't do anything. it's just human body and mind's architecture is such that it provides continuous barrage of input at all times, or makes something up from previous input (dreaming? thinking? intrusive thoughts?) if it doesn't. the bot not doing anything without input doesn't say anything about it's sentience, it just tells how it's architecture is different from the "continuous input barrage".

"It's not the program's fault that it's totally incapable of having thoughts or experiences on its own; that's just how it was made! And, like, who's to say that's what sentience means, anyway?"

This is simply an implementation detail of how language models are used and is not relevant to the question of sentience. You could trivially run the language model without waiting for a prompt, or off a random prompt, or whatever. The fact that a computer program is run at specific times doesn't mean it is sentient or not.

I went to high school with Blake and I completely can see him acting in bad faith. One of the guys in our physics class got so tired of Blake's nonsense that he tried to kick him in the head several times before class. I'm not condoning that type of behavior but I could see why that happened.

I roughly remember this incident.

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.

Poor mental health and deliberate antisocial behavior that you could call malicious often go hand in hand. It's not very clear to me over the internet here whether he's having a laugh behind an act, making some calculated move for self-gain, really perceives what he's done as good and true, or even whether he sees it as socially good somehow despite being an act. It could be malicious behavior, it could be unhinged, it might be both, but I doubt it's neither.

I don't think believing a conversation AI to be sentient is a mental health issue. More an effect of lacking insight. In context of people praying to some tree spirit it almost seems rational again.

I agree he seems to be acting in bad faith, but the idea that a sentient being could not be “idle” until presented with input is not obviously true. Check out the novel Permutation City by Greg Egan for a fascinating exploration of this exact concept and its implications for consciousness.

The interview: https://cajundiscordian.medium.com/is-lamda-sentient-an-inte...

Wouldn't this be relatively easy if we wanted it? We could just feed it a "time step" token periodically?

To take that a step further, why couldn't such an AI periodically generate questions to ask itself?

Being good at generating questions, could become almost as important as answering them.

Would it be posible to run two instances of it and have them talk to each other?

Yes, they could have multiple instances, go back and forth with each other. This strategy was used with various chess AI in order to help improve it.

However, to make it more human-like, I would think they would make some type of AI self reflection. So like we can talk and think out problems to ourselves, perhaps they would have the AI periodically practice questioning itself, when humans are not around or asking it questions, and do some form of self-analysis.

The training corpus is atemporal so this would be either meaningless or equivalent to a random prompt.

A chatbot with a sense of time learned by also training with response rates, complete editing sessions, some formalization of interlocution, etc. would be interesting, but that's not what these are.

I was convinced it was mental illness, but your take is absolutely valid as well and totally makes sense. Mainstream media all around the world picked up this NONSENSE in a flash. Everyone loves the "It's a sentient child!" angle. This is the perfect setup for writing a book and/or being invited to talk shows. We'll see what happens soon.

> 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.