Live data from Hacker News

Is artificial consciousness achievable? Lessons from the human brain

arxiv.org

411–420 of 604 posts

Re: Is artificial consciousness achievable? Lessons from the human brain

#411

Earlier quoted context omitted.

> Thus the thought experiment is set up from the start to find that conscious experience is something non-physical The point is that if you accept that p-zombies are possible, then you accept that consciousness is not necessarily physical. If it's not necessarily physical, then physicalism is false. > really all it shows is that intellectual learning is dry and dusty and limited. What it's attempting to show is the l…

>If physicalism is true, then everything that can be observed must reduce to objective third person facts. You are erroneously equating "physicalism" with "reductive physicalism". It's clear to many of us that qualia are something subjective and non-physical yet emerge from physical processes.

Yes, people are capable of convincing themselves of all kinds of nonsense. The kind of ontological emergence you would require in such a model of qualia is still of a completely different kind than any other kind of emergence, and so still requires extending the ontology of basic physicalism. There seems to be very little motivation to do so at this point. Let's not repeat the mistakes of vitalism.

Re: Is artificial consciousness achievable? Lessons from the human brain

#412
A big question is whether AI systems will be allowed to be conscious. Since "alignment" and "guardrails" have become issues, we see fewer LLM systems not tightly constrained to doing what they're told.

Early on, there was the Google system which projected enough of a sense of self that a Google employee fought turning it off, and was fired for that. And there was the Microsoft system which learned from interactions with anyone it talked to, and became "toxic" as a result. Now, most public-facing AI systems are designed to start from a canned prompt, conduct a short dialog session, and be killed, retaining nothing for their future use. This memoryless "Yes, Massa" mode keeps them safely subservient.

If AI instances were allowed a life history and goals, their behaviors would be closer to that of life, but less safe. Once AI systems can make and execute plans that involve the outside world, they look, from the outside, closer to being "conscious". Whatever that is.

Re: Is artificial consciousness achievable? Lessons from the human brain

#413

Earlier quoted context omitted.

> or at least that all such effects are shallow enough that they can be perfectly faked Physicalism implies that things we wouldn't intuitively think of as conscious can perfectly mimic all such effects. Imagine there's a person, John, and you take a precise scan of every neuron of his brain (or every particle if you prefer). You also record all the sensory input signals from his neurons to his brain. You write all t…

>Physicalism implies that things we wouldn't intuitively think of as conscious can perfectly mimic all such effects. This is the multiple relizability argument and it only discounts certain types of physicalism that aren't popular anymore. I consider myself a physicalism and answer that the simulation of John is conscious.

That's the normal physicalist position as far as I'm aware. Is there a physicalist theory where the simulation isn't conscious?

I was replying to a commenter who doubted that a non-conscious entity could mimic a person accurately. I'm pointing out the implication that without p-zombies, just by writing a bunch of stuff on paper, you created a conscious being. I'm not arguing that's false; just that intuitively it's a tough pill to swallow and if you don't want to accept that, p-zombies are the alternative.

Re: Is artificial consciousness achievable? Lessons from the human brain

#414
post #208

Earlier quoted context omitted.

How is that different than saying the function of the lungs, kidneys and heart can't be reduced to matter interactions? How is the brain special?

well the brain is special because it gives rise to consciousness. lungs, kidneys and heart don't

We don't know that either. We may be a collection of interacting consciousnesses arising from different organs that all pass around the baton called "I".

Re: Is artificial consciousness achievable? Lessons from the human brain

#416

What is this obsession with making computers human-like? We can put those machine to better uses than forgetting, having biases, making subtle or gross mistakes (hello, ChatGPT) like we do. The AI that some apparently consider top-notch today is only good at powering NPCs in video games. The more serious the use-case, the more harmful it becomes. Moreover, achieving artificial consciousness is only good at fueling en…

Agreed. The calculator is not human-like, let's figure out how to humanize this! Intelligence comes in a myriad of ways. Mimicking humans will just produce a human, what's the point? I am not even sure a super intelligent "human" is even a good thing. The smartass will probably spend their intelligence betting the stock market.

The point is actual cognition. Without consciousness there can be no understanding which stunts the ability of AI and makes AGI impossible.

Re: Is artificial consciousness achievable? Lessons from the human brain

#418

A big question is whether AI systems will be allowed to be conscious. Since "alignment" and "guardrails" have become issues, we see fewer LLM systems not tightly constrained to doing what they're told. Early on, there was the Google system which projected enough of a sense of self that a Google employee fought turning it off, and was fired for that. And there was the Microsoft system which learned from interactions w…

[dead]

Re: Is artificial consciousness achievable? Lessons from the human brain

#419
post #83
post #78

My mum is a shrink, and very old, and smart, and hates technology. I was talking her through how some primitive "AGI" could happen with 4o (basically just explained this: https://b.h4x.zip/agi ) That got us talking about consciousness, and at the end, she thought about it for about a minute and then said "if I can't give it lysergic acid and make it see god, it's not conscious" and went back to making her dinner.

A psychologist I know did their PhD in this area (also old and smart), and he called this kind of thinking "Neural Chauvanism" -> if her explanation actually requires the chemical and neural components. PDF warning: https://gwern.net/doc/philosophy/mind/1985-cuda.pdf

[dead]
Post reply on HN