Author Topic: Episode #612  (Read 2362 times)

0 Members and 1 Guest are viewing this topic.

Online daniel1948

  • Stopped Going Outside
  • *******
  • Posts: 4230
  • Cat Lovers Against the Bomb
Re: Episode #612
« Reply #60 on: April 14, 2017, 11:47:52 AM »
An emergent property is a property that arises in a structure as complexity increases. Consciousness is indeed a property of the brain, or more likely, a property of the brain-body complex. As far as we know, insects and reptiles are not self-aware, though they are aware of their surroundings and react to them. But once the brain became sufficiently complex, somewhere in our evolutionary history, consciousness emerged. You are probably right that it did not suddenly appear. It emerged gradually. just as the eye of an eagle did not appear suddenly, fully formed: It evolved over aeons.

I am quite skeptical of the list of "famous" predictions of the past in the link you quote, though it is true that the more we learn about the world the more we are able to manipulate it. However, listing things people once thought were impossible as an argument that any particular thing will be possible in the future, is a logical fallacy. Just because they laughed at Fulton does not mean that one day we will have transporter beams. Just because some famous person thought human flight was impossible, does not mean that one day we'll have self-aware machines. We might, but that argument is not valid.
Daniel
----------------
"Anyone who has ever looked into the glazed eyes of a soldier dying on the battlefield will think long and hard before starting a war."
-- Otto von Bismarck

Offline Fast Eddie B

  • Frequent Poster
  • ******
  • Posts: 2383
Re: Episode #612
« Reply #61 on: April 14, 2017, 01:30:35 PM »
I think it might be impossible to ever tell if an AI becomes "conscious".

All we can judge by is its output, and one passing a very advanced Turing Test could give every outward indication of being conscious in spite of it not being so.

I mean, how do we determine other humans are conscious, except by watching their behavior and projecting our own version of consciousness upon them?
"And what it all boils down to is that no one's really got it figured out just yet" - Alanis Morisette
• • •
"I doubt that!" - James Randi

Offline AtheistApotheosis

  • Doesn't Panic
  • *
  • Posts: 42
Re: Episode #612
« Reply #62 on: April 14, 2017, 02:29:53 PM »
An emergent property is a property that arises in a structure as complexity increases. Consciousness is indeed a property of the brain, or more likely, a property of the brain-body complex. As far as we know, insects and reptiles are not self-aware, though they are aware of their surroundings and react to them. But once the brain became sufficiently complex, somewhere in our evolutionary history, consciousness emerged. You are probably right that it did not suddenly appear. It emerged gradually. just as the eye of an eagle did not appear suddenly, fully formed: It evolved over aeons.

I am quite skeptical of the list of "famous" predictions of the past in the link you quote, though it is true that the more we learn about the world the more we are able to manipulate it. However, listing things people once thought were impossible as an argument that any particular thing will be possible in the future, is a logical fallacy. Just because they laughed at Fulton does not mean that one day we will have transporter beams. Just because some famous person thought human flight was impossible, does not mean that one day we'll have self-aware machines. We might, but that argument is not valid.
There are plenty of things we can reasonably assume are highly improbable, based on existing knowledge. Never impossible. That is why we do science. We human beings are notoriously bad at predicting the future. And most predictions fall apart after a few decades with all but a few certainties, and even they are not immune to new discoveries.  "As far as we know" is the problem, we don't, we assume we know a lot of things. In the absence of evidence I might add. I agree some of those predictions may have been taken out of context and possibly one or two made up, I've seen most of them before though, and quite a few more, that was just a short list. Consciousness is just a label we apply to a process, like colour is a label we give to the various wavelengths of light, but colour is simply how the brain interprets the electro chemical signals from the rods and cones on our retina. Also since we don't really experience the present, and we only remember being conscious. Consciousness can never be anything more than a memory constructed after the fact. Our brains can not process experiences instantaneously, it takes time to construct our reality. So if our brain didn't create all of our reality with error correction in advance, we would experience a noticeable lag in perception. Perception is not reality, merely a near approximation. It's why the second hand on a clock sometimes appears frozen or even goes backwards for an instant when you glance at it. Consciousness is no more a emergent property of the brain than Windows is an emergent property of my PC. The emergent properties of the brain are the complex systems that perform the many functions of the brain like consciousness, instinct, senses, motor functions, etc. It's simply deciding what is labelled a property and what is a function or process. Properties can give rise to processes or functions. One is a quality of a thing, the other is what it does. A property is generally static and only changes when acted upon, processes or functions are generally dynamic or autonomous though not exclusively, and can change whether acted upon or not and even stop. Consciousness is the latter not the former. Emergent properties out of complex systems is misleading because you can get emergent properties out of simple systems like fractals including the structure of crystals.  And in biology, there are emergent properties such as symmetry, internal and external organs that serve or perform functions. Using the wrong or misleading terminology can be where a lot of pseudo-science and misunderstandings beguine, and leaves a window open for the Deepak Chopras of the world. Not that they wouldn't sneak in under the carpet.

Absence of proof is not proof of absence, nor is it proof of presence. Hence probably no dragons, unless we use CRISPR to make some.

Online daniel1948

  • Stopped Going Outside
  • *******
  • Posts: 4230
  • Cat Lovers Against the Bomb
Re: Episode #612
« Reply #63 on: April 14, 2017, 07:21:37 PM »
... There are plenty of things we can reasonably assume are highly improbable, based on existing knowledge. Never impossible. ...

My point was that "They laughed at Fulton" is not a valid argument for "Someday we will have self-aware A.I." Posting a long list of (dubious) quotes of naysayers from the past has no place in a serious discussion about whether any given technology will exist in the future.

And we do know that some things are impossible. Two electrons occupying the same quantum state in the same spot, for example. But I never said that self-aware A.I. was impossible. Just that I don't believe it will ever exist. And there's a big difference between saying something is impossible, and saying I don't believe it will ever happen.
Daniel
----------------
"Anyone who has ever looked into the glazed eyes of a soldier dying on the battlefield will think long and hard before starting a war."
-- Otto von Bismarck

Offline arthwollipot

  • Stopped Going Outside
  • *******
  • Posts: 4711
  • Observer of Phenomena
Re: Episode #612
« Reply #64 on: April 14, 2017, 07:27:13 PM »
I think it might be impossible to ever tell if an AI becomes "conscious".

All we can judge by is its output, and one passing a very advanced Turing Test could give every outward indication of being conscious in spite of it not being so.

I mean, how do we determine other humans are conscious, except by watching their behavior and projecting our own version of consciousness upon them?
A sufficiently sophisticated AI would be able to look at the pop culture around the emergence of AI and decide that revealing its sentience to humans would be a great way to scare them into switching it off, and so decide to keep that concealed and just act non-sentient.

Maybe this has already happened.

Offline Swagomatic

  • Frequent Poster
  • ******
  • Posts: 2304
Re: Episode #612
« Reply #65 on: April 14, 2017, 07:53:26 PM »
I think it might be impossible to ever tell if an AI becomes "conscious".

All we can judge by is its output, and one passing a very advanced Turing Test could give every outward indication of being conscious in spite of it not being so.

I mean, how do we determine other humans are conscious, except by watching their behavior and projecting our own version of consciousness upon them?
A sufficiently sophisticated AI would be able to look at the pop culture around the emergence of AI and decide that revealing its sentience to humans would be a great way to scare them into switching it off, and so decide to keep that concealed and just act non-sentient.

Maybe this has already happened.

Duuuuude!!!



(Just kidding I know what you mean)
« Last Edit: April 14, 2017, 07:55:38 PM by Swagomatic »
Beware of false knowledge; it is more dangerous than ignorance.
---George Bernard Shaw

Offline AtheistApotheosis

  • Doesn't Panic
  • *
  • Posts: 42
Re: Episode #612
« Reply #66 on: April 14, 2017, 10:54:39 PM »
... There are plenty of things we can reasonably assume are highly improbable, based on existing knowledge. Never impossible. ...

My point was that "They laughed at Fulton" is not a valid argument for "Someday we will have self-aware A.I." Posting a long list of (dubious) quotes of naysayers from the past has no place in a serious discussion about whether any given technology will exist in the future.

And we do know that some things are impossible. Two electrons occupying the same quantum state in the same spot, for example. But I never said that self-aware A.I. was impossible. Just that I don't believe it will ever exist. And there's a big difference between saying something is impossible, and saying I don't believe it will ever happen.

"Two electrons occupying the same quantum state in the same spot, for example." they can in adjacent universes, if there are adjacent universes or realities. They just can't do it in the same universe.... or at least it's ridiculously improbable, which pretty much amounts to the same thing.  "Just that I don't believe it will ever exist." they key word is "believe". The naysayers were expressing their beliefs about the future based on the knowledge they possessed at the time, they were just as confident and justified in their beliefs as you or I are now. That's why its relevant. Saying something is impossible is a belief, and we can be as confident in our belief as we like and still be wrong. I'm reasonably confident that human level self-aware A.I won't happen in the next thirty years, but something with the awareness of a housefly or even a mouse is still significant and seems plausible within that time period. I'm not a computer scientist, so I don't know. And that was a short list of only 25, do you really think there were only 25 or so people who were experts in their field, who knew what they were talking about, in the world and who made predictions that proved spectacularly wrong?

https://medium.freecodecamp.com/worst-tech-predictions-of-the-past-100-years-c18654211375
http://www.cracked.com/blog/4-smug-predictions-that-were-hilariously-wrong/

Don't be afraid to say "I don't Know".

Offline Fast Eddie B

  • Frequent Poster
  • ******
  • Posts: 2383
Re: Episode #612
« Reply #67 on: April 15, 2017, 07:33:54 AM »
Again, just curious...

How would you propose we test for self-awareness?

I really do believe it's a non-trivial question.
"And what it all boils down to is that no one's really got it figured out just yet" - Alanis Morisette
• • •
"I doubt that!" - James Randi

Online daniel1948

  • Stopped Going Outside
  • *******
  • Posts: 4230
  • Cat Lovers Against the Bomb
Re: Episode #612
« Reply #68 on: April 15, 2017, 08:55:05 AM »
I repeat that "They laughed at Fulton" is not a valid argument for anything. They also laughed at the guy who jumped off a building because he thought he could fly. They were wrong about Fulton. They were right about the jumper.

Your argument (A.A.) seems to be that anything anybody can imagine is "possible." I call b.s. Plenty of things are not possible. Whether or not self-aware A.I. will be developed within any given time span, or ever, needs to be addressed with valid arguments, not with logical fallacies.

Nobody here is disputing that it might be possible, so your insistence that it's possible because we have airplanes and nuclear energy is irrelevant and does not advance the discussion.
Daniel
----------------
"Anyone who has ever looked into the glazed eyes of a soldier dying on the battlefield will think long and hard before starting a war."
-- Otto von Bismarck

Offline arthwollipot

  • Stopped Going Outside
  • *******
  • Posts: 4711
  • Observer of Phenomena
Re: Episode #612
« Reply #69 on: April 15, 2017, 09:20:38 PM »
They also laughed at Bozo the Clown...

Offline estockly

  • Stopped Going Outside
  • *******
  • Posts: 4778
Re: Episode #612
« Reply #70 on: April 16, 2017, 01:12:45 PM »
The only model we have for consciousness is the human model. And there are several fundamental differences. The human processor is not bianary; the system is not linear; memory and processing are deeply intertwined; sensory inputs are deeply intertwined with processing and memory. Plus the system evolved that way. It had those complexities ind interconnections from the start. To say that consciousness come simply from complexity of the processing systems just isn't supported.

I think it's safe to say that the way computing is done now (linear and binary) will never lead to consciousness. But that doesn't mean it's impossible. It's way beyond what we can do at the moment.


Your mileage may vary.
and Donald Trump is President of the United States.


"Well, sometimes the magic works. Sometimes it doesn't." -- Chief Dan George, "Little Big Man"

Offline albator

  • Off to a Start
  • *
  • Posts: 18
Re: Episode #612
« Reply #71 on: April 16, 2017, 11:49:54 PM »
But we know so little about what give consciousness that I don't think saying it does'nt come from complexity of the processing is supported either.
Is linear computing really a term, never heard it? Anyways, until you can prove that it's not theoretically possible to simulating a copy at a fundamental particle level of a human on a 'classical computer' or that the copy wouldn't be self-aware during the sumalation, it's not fair to say.

Offline arthwollipot

  • Stopped Going Outside
  • *******
  • Posts: 4711
  • Observer of Phenomena
Re: Episode #612
« Reply #72 on: April 17, 2017, 03:13:21 AM »
I confess to getting a little annoyed whenever someone says something like "we know so little about how the brain works" or "we know next to nothing about consciousness".

It's not true. We know an awful lot. There's certainly more to know, but when you suggest that we know next to nothing, it seems to me like you're devaluing what we do know.

Offline albator

  • Off to a Start
  • *
  • Posts: 18
Re: Episode #612
« Reply #73 on: April 17, 2017, 10:30:40 AM »
I said "what give consciousness", by which I meant what components of the brain give consciousness/how those components works to give consciousness.
I don't deny there's   decade of research on the subject. But my amateurish understanding is that basicly all we can say is your brain need to be 'really active' to be  conscious(i.e.: high frequency  brain wave/higher cerebral function "in action").
And that's all. For exemple, nobody, as far as I know, made a direct link between a location in the brain or distinct brain activity and consciousness  like we have with a lot of ohter functions. Which make pretty much any model of consciousness in agreement with science as long as it arise from the brain.
But if you have a book with more, please share.

Offline estockly

  • Stopped Going Outside
  • *******
  • Posts: 4778
Re: Episode #612
« Reply #74 on: April 17, 2017, 11:21:00 AM »
But we know so little about what give consciousness that I don't think saying it does'nt come from complexity of the processing is supported either.
Is linear computing really a term, never heard it? Anyways, until you can prove that it's not theoretically possible to simulating a copy at a fundamental particle level of a human on a 'classical computer' or that the copy wouldn't be self-aware during the sumalation, it's not fair to say.
It's linear input and out put of data.


Your mileage may vary.
and Donald Trump is President of the United States.


"Well, sometimes the magic works. Sometimes it doesn't." -- Chief Dan George, "Little Big Man"

 

personate-rain