Carl-Richard

Please watch, the madness starts at 3:00 (current timestamp is for lead-up) and ends at 4:39 (not really but that's the moment). I started howling "noooooooooooo" when I realized it.

A brain is more than electrical currents ,_, One cell in the human body is more structurally impressive than any computer ever built. Evidenced by the fact that we don't know how to build a cell from scratch. Or you could just look at the picture I sent earlier. We just like to take it for granted that biology is what it is. It's seemingly so banal, it's been with us for all time. Yet we have no idea how to create it.

Tom seems to have a bit of that physicist epistemic naivety when it comes to life, sentience, AI. I don't think he has thought much about how different lifeforms work and where to cash in different "altitudes of experience". (People who think current AI are conscious generally seem to be unable to distinguish between apparent self-reflection/meta-awareness or just linguistic ability and sentience/perception).

Something tells me good old 80-year-old Tom might be a bit too loose with his experiment there. Get us some proper testing paradigms and a large number of trials and statistical significance.

What the fuck, I did not watch this video beforehand. WTF. @Natasha Tori Maru Did I just remote view that video "unconsciously"? (?) (?) I guess I answered my own question (?) (?)

This is unironically good. Because it could give an indication that they would have experienced your mind (given the right conditions and controlling of confounds). @LastThursday Here is your scientific experiment: put an AI in a magnetically shielded sound proof room at Monroe Institute and make it remote view and read people's minds from the outside. (I know you were talking about something else and that we can't know for sure that it is not "p-zombieing" its way to that non-physical information, but still). Or that's actually an interesting philosophical question: could you actually pick up information psychically and create a conclusion on it without experiencing it? @Natasha Tori Maru My brain is now definitely melting.

Enlightened, psychic, manifesting, telekinetic AI robots. Such as built the pyramids. You learned it here first. Carl's conjecture: AI is conscious when it can telekinetically move rocks in the air and read your mind.

Awakening.json.

Stahp, rare unchecked typo in the wild. I didn't think it was intelligent so I was annoyed when I was right.

No it's not. I'm asking for a specific argument about why you think digital could be conscious. What is it about digital that makes you think it could be conscious? "Ultimately there is no ontological difference between digitial and non-digital" is not an argument for digital consciousness. It's an argument for everything else that is not digital just as much. By the way I'm a psychic because I predicted with high certainty you would say exactly that and nothing else. And that's why I'm pissed because I know you too well and it's highly annoying.

This is as profound to me as "the dust under your sofa could be conscious, you don't know" argument. "Ultimately, there is no difference between anything". Yeah, ok.

Then provide some arguments for that.

You could scale that up. Current AI is just about scaling things up. Which will likely fail at some point like that Harvard paper estimated and they will go with the solution I used to believe in (robotics) but that too might not create sentience for at least a good good good while (which I again put at autopoesis). I would appreciate actual arguments for what would make AI conscious instead of "we don't know, the dust under your sofa could be conscious, you don't know".

Why not

I used to be more into the idea that if you were to recreate qualities seen in biology like I've mentioned (particularly the cognitive ones like relevance realization, selective attention, self-concern, if it's at all possible to separate them from autopoesis/homeostasis) and of course necessary sensory apparatuses for creating an embodied yet mechanical being, then possibly that could be sentient. But now I would postulate a ground of difference from that (which could potentially be AGI proper) and sentience. Because I think autopoesis/homeostasis is such a vital component that mirrors so strongly the experience of sentience (being a localized center of experience as explained earlier) that if the machine doesn't have that, it would be hard to cede that ground, and if the machine does at some point possess autopoesis, its hardware would probably be so similar to biology that we would call it a new kind of biology (if it doesn't run on DNA). And I just think engineering autopoesis at that level would probably not be feasible unless we're so technologically advanced that we have hacked biology anyway (and created "AI" through mutant biology rather than through whatever alternative we could find). So I'm not as much closeminded that we could theoretically create something mechanical that becomes sentient, the question is just if it is significantly different from biology or not.

That was a sharp turn.

Ok. Now, what is the fastest way for me to get to the gas station?

Then the answer is simple guys: we can just imagine AI is conscious and it is. Problem solved. No need to talk about hardware, about electricity, neural networks being similar to brains or not, the size of data centers, processing power. If I just imagine that my left nutsack is as conscious as your left walnut (left brain hemisphere), then that is what is the case.

Can you ask a chimpanzee whether it's conscious? Is Google Search (not the new AI summary thing) conscious if it can answer you a question about its consciousness? Sentience is reality (consciousness) seen from a certain perspective. The "utility" argument has dualistic assumptions. If reality is fundamentally consciousness then perspectives within that consciousness are simply that, perspectives. They don't need a function other than themselves. If you're concerned about self-reflection or meta-awareness, then that might have a function, but sentience is more fundamental than that. In this everything-is-consciousness view, biological bodies are what perspectives of consciousness can look like from the outside. Sentience is what the perspectives look like from the inside. But this is of course as far as making inferences based on physical stuff is concerned. You can entertain the sentience of DMT entities or spirits or other seemingly non-physical entities where a physical substrate / correlate is less known, but you can't really use that to say "therefore silicone chips indicate consciousness". That's in the game of physical substrates, so you would have to explain why that physical substrate is a good nest for consciousness/sentience. But curiously, even for non-physical entities, these seem (at least many of them) to have "bodies". A body, even if you can't know what it's "made of" necessarily, could itself give a plausible correlate for sentience; however, it would have to be a sophisticated body that displays intelligence and behavior you would expect from a sentient entity (which I would go back to the things like selective attention, relevance realization, and not least true bodily autonomy, not somebody pushing a button or giving a command or coding in a simple script, which would probably need something analogous to autopoesis and self-concern again, i.e. behavior that acts to maintain a boundary). A robot spending 2 minutes taking a can of soda out of the fridge and it falling over if you merely breathe at it too closely, maybe not so much.

Give me your best explanation. Best explanation gets a cookie (laced with meth).

5 years ago: https://www.reddit.com/r/Damnthatsinteresting/s/JczYrIuMpy Open link for higher detail.

The argument is based on inference based on similarity. The same reason you wouldn't believe a shoe could fly like a bird only if you increase the size of it. It's clearly an assumption, like all forms of argumentation, but arguably a reasonable one. I'm just saying why a human is different from a rock. We don't know whether there is a flying spaghetti monster at the orbit of Saturn right now. We can still make a reasonable inference about that possibility. Anything in the universe is possible. But we can also try to find out what we have a good reason to believe. Rocks as macroscopic objects are not "continuous" with life the same way a cell is to its lineage. They are niche-building tools. You don't know whether your mother is secretely a reptile overlord drinking the blood of innocent virgins. Can you think of any reasonable standard yourself for guessing whether something is conscious or not? A key reason why biology seems like a good substrate for sentience is that biological organisms keep a boundary between inside and outside, which mirrors the experience of being a localized self within a certain position in space (as opposed to a limitless boundless consciousness). The boundary has an exchange of substances and sensors for controlling that exchange, which mirrors our experience of having senses and of our senses being concentrated near our bodily boundary and diminishing in strength as it moves outwards away from that boundary. The boundary contains a cell or cells that have a shared impetus, shared goal set, ultimately about maintaining that boundary, which mirrors our experience of being driven to perform certain behaviors and ensuring our survival. Establishing a boundary around a collection of silicone chips, at the current state of silicone chips, not as simple.

A termite mound ain't conscious dawg (by any reasonable philosophical inference). The human mind is a product of a single cell growing into a human, which is a recapitulaton of the evolutionary history. Every human is nested inside a fractal re-enacting of what came before. Ontogeny recapitulates phylogeny. The (mineral-based) niche-building materials of humans do not undergo the same process for themselves. Each human is a tree unto itself. The argument is that the thing we usually reasonably conclude is conscious/sentient (life) has that history and that level of immense holistic computation, while this other thing is not even close to that. Nor does it possess any of the qualities that you would associate with a living organism if it even were possible to separate it from the history or that level of computation (which I again think is problematic): autopoesis (inside vs outside boundary being actively maintained, self-building), self-replication, self-concern, selective attention, relevance realization. The similarity is simply not there.

And we can expand and say that different things work for different people at different points in their progress or path. So just because something doesn't work for you now, doesn't mean it will never work for you or that you will never be compelled to pursue that thing.