--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAJPayv2v2BPqSbKyDSRiRCYAs9jAVRMtEAA5EfscX8nukJzkEw%40mail.gmail.com.
> It's certainly impressive...and useful. But notice how very different it is from what we call intelligence in humans.
> I wonder what kind of consciousness you would infer from DeepMind's behavior?
This is indeed a historic moment for AI - protein folding is unbelievably complex and to now have a tool that can deal with that complexity is of inestimable value. But I do have these concerns:
- It will be tempting to assume that DeepMind is correct on any given structure. But we don't have any easy way to test it. Of course, we can have a high degree of confidence that the predicted shape is accurate, and the value in that is already huge. But mistakes will be made based on this assumption.
- This tool can be weaponized to create new and even highly-targeted poisons. It's not hard to imagine developing a poison that was only toxic for people of a certain race and then delivering it via virus.
- Who has access to DeepMind?
- Are we comfortable with a corporation controlling something so powerful and with potential global security issues? This question will only get increasingly more relevant as new advances in AI are made. Can the world ever hope to regulate something so simultaneously powerful and cutting edge?
Terren
--On Sun, Jul 25, 2021 at 8:56 AM John Clark <johnk...@gmail.com> wrote:
--In my opinion this is the most impressive thing that Artificial Intelligence has done to date:From The New York Times:
A.I. Predicts the Shapes of Molecules to Come
DeepMind has given 3-D structure to 350,000 proteins, including every one made by humans, promising a boon for medicine and drug design.
https://www.nytimes.com/2021/07/22/technology/deepmind-ai-proteins-folding.html?smid=em-share
John K Clark
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAJPayv2v2BPqSbKyDSRiRCYAs9jAVRMtEAA5EfscX8nukJzkEw%40mail.gmail.com.
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAMy3ZA9dE9dRujPyEXjsyf3NaKHFJaSDvCvJ4q5iAWsAnUfcdQ%40mail.gmail.com.
On Sun, Jul 25, 2021 at 2:43 PM 'Brent Meeker' via Everything List <everyth...@googlegroups.com> wrote:
> It's certainly impressive...and useful. But notice how very different it is from what we call intelligence in humans.
At least in this case the biggest difference is the artificial version of Intelligence works one hell of a lot better than the non-artificial human variety.
> I wonder what kind of consciousness you would infer from DeepMind's behavior?
I've always thought intelligence is hard but consciousness is easy, so to be consistent I'd have to say it must possess consciousness of some sort. Subjectively I have no way of knowing what it would feel like to be DeepMind, but then subjectively I have no way of knowing what it would feel like to be Brent Meeker either.
John K Clark See what's on my new list at Extropolis
vqmz
DeepMind has given 3-D structure to 350,000 proteins, including every one made by humans, promising a boon for medicine and drug design.
https://www.nytimes.com/2021/07/22/technology/deepmind-ai-proteins-folding.html?smid=em-share
John K Clark
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAJPayv3bNdcU3xy2CzppTpY1DwJRScHE9mtzKaUrTY5v6rAajw%40mail.gmail.com.
> And you have no way of knowing what it will feel like to be John K Clark tomorrow, but you have a pretty good theory about it.
> Similarly, you probably have a better theory about what it would feel like to be Brent Meeker than to be DeepMind.
> Consciousness is imagined be an impossibly hard problem because it's posed as being able to predict conscious thoughts from monitoring a brain.
> But that's like saying gravity is a hard problem because we can't predict the motion of all the stars in a galaxy (or even three bodies).
> the fact that you can say the consciousness of DeepMind might be so different you have no way of knowing what it would be like implies that there can be qualitatively different kinds of consciousness.
> the fact that you can say the consciousness of DeepMind might be so different you have no way of knowing what it would be like implies that there can be qualitatively different kinds of consciousness.
Yes. I only have experience with my own consciousness but I know for a fact that depending on the time of day my consciousness can be qualitatively different, and I've known that for a long time. Back when I was a student taking a calculus exam my consciousness had reached a high-level but later that same night when I was falling asleep it was at a much lower level and just a little later it fell all the way to zero, and then what seemed instantaneous but actually took 8 hours it started up again.
>> I only have experience with my own consciousness but I know for a fact that depending on the time of day my consciousness can be qualitatively different, and I've known that for a long time. Back when I was a student taking a calculus exam my consciousness had reached a high-level but later that same night when I was falling asleep it was at a much lower level and just a little later it fell all the way to zero, and then what seemed instantaneous but actually took 8 hours it started up again.
> True. But you left off the interesting conclusion,
> i.e. that different kinds of consciousness maybe connected to different kinds of intelligence.
> Just as your consciousness can go from awake to asleep (which is actually different from unconscious), it can also be merely perceptive, or it can imagine things, or think of a narrative story, and these can be mixed with various emotional feelings. Right?
> If you agree or not, either way it implies that we can test theories of consciousness.
In the final analysis the only way to test a theory is by making objective observations about the way things behave, but consciousness is a subjective phenomenon and that's what causes the problem. The theory that other humans besides me are conscious is perfectly consistent with all observable evidence, but so is the theory that I am the only conscious being in the universe,
and so is the theory that EVERYTHING is equally conscious, even grains of sand, even atoms, even quarks and electrons. The trouble is ANY consciousness theory will fit the observable facts just fine, and that's why ALL consciousness theories are utterly useless, except for the theory that solipsism is untrue, that one has a use.
On 7/25/2021 3:39 PM, John Clark wrote:
In the final analysis the only way to test a theory is by making objective observations about the way things behave, but consciousness is a subjective phenomenon and that's what causes the problem. The theory that other humans besides me are conscious is perfectly consistent with all observable evidence, but so is the theory that I am the only conscious being in the universe,
But that's inconsistent with the theory that consciousness is instantiated by physical processes in the brain. And that theory is supported by many observations and experiments on brains and the reports by subjects.
and so is the theory that EVERYTHING is equally conscious, even grains of sand, even atoms, even quarks and electrons. The trouble is ANY consciousness theory will fit the observable facts just fine, and that's why ALL consciousness theories are utterly useless, except for the theory that solipsism is untrue, that one has a use.
But that's the way all theories are. We provisionally believe the ones that are consistent with the facts and are most useful. And the theory of minds, that other people (and animals) are conscious and have an internal narrative, is extremely useful and in fact any human lineage that did not hold that theory has already been eliminated by evolution. Did you ever read Julian Jaynes "The Origin of Consciousness in the Breakdown of the Bicameral Mind"?
Brent
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/c732b029-b833-6eb4-f4d0-ed366fa90003%40verizon.net.
On Sun, Jul 25, 2021 at 4:44 PM 'Brent Meeker' via Everything List <everyth...@googlegroups.com> wrote:> And you have no way of knowing what it will feel like to be John K Clark tomorrow, but you have a pretty good theory about it.Yes, and tomorrow I will be able to definitively know if yesterday's theory about what it will be like to be John K Clark today turned out to be correct or not.> Similarly, you probably have a better theory about what it would feel like to be Brent Meeker than to be DeepMind.No, there's nothing similar at all about it because tomorrow I will STILL have absolutely positively no way of knowing if yesterday's theory about what it will be like to be Brent Meeker or DeepMind today turned out to be correct or not, in fact I will NEVER know if it's correct> Consciousness is imagined be an impossibly hard problem because it's posed as being able to predict conscious thoughts from monitoring a brain.The hardest part of the "hard problem of consciousness" is clearly explaining exactly what "the hard problem of consciousness" is, it's not at all clear to me exactly what sort of explanation would satisfy the consciousness gurus.
> But that's like saying gravity is a hard problem because we can't predict the motion of all the stars in a galaxy (or even three bodies).I can make exact Newtonian predictions in a few very special situations but in general you're right, I can't make an exact prediction of the motion of 3 particles, but I can make some very good approximations, and by using The Virial Theorem I can even make a good approximation for the motions of millions of bodies. However I don't know, and will never know, if my predictions about a consciousness other than my own is even approximately correct. And that's why consciousness theories are so easy to dream up, and that's also why they're such a colossal bore.> the fact that you can say the consciousness of DeepMind might be so different you have no way of knowing what it would be like implies that there can be qualitatively different kinds of consciousness.Yes. I only have experience with my own consciousness but I know for a fact that depending on the time of day my consciousness can be qualitatively different, and I've known that for a long time. Back when I was a student taking a calculus exam my consciousness had reached a high-level but later that same night when I was falling asleep it was at a much lower level and just a little later it fell all the way to zero, and then what seemed instantaneous but actually took 8 hours it started up again.John K Clark See what's on my new list at Extropolis0o0o
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAJPayv2mxNZd8m_P4Z6ggdV2sgmV3BmjC8DGViuLacwO8ZVoJQ%40mail.gmail.com.
>> In the final analysis the only way to test a theory is by making objective observations about the way things behave, but consciousness is a subjective phenomenon and that's what causes the problem. The theory that other humans besides me are conscious is perfectly consistent with all observable evidence, but so is the theory that I am the only conscious being in the universe,> But that's inconsistent with the theory that consciousness is instantiated by physical processes in the brain.
> And that theory is supported by many observations and experiments on brains and the reports by subjects.
>> and so is the theory that EVERYTHING is equally conscious, even grains of sand, even atoms, even quarks and electrons. The trouble is ANY consciousness theory will fit the observable facts just fine, and that's why ALL consciousness theories are utterly useless, except for the theory that solipsism is untrue, that one has a use.
> But that's the way all theories are. We provisionally believe the ones that are consistent with the facts
> And the theory of minds, that other people (and animals) are conscious and have an internal narrative, is extremely useful and in fact any human lineage that did not hold that theory has already been eliminated by evolution.
> Did you ever read Julian Jaynes "The Origin of Consciousness in the Breakdown of the Bicameral Mind"?
On Sun, Jul 25, 2021 at 7:09 PM 'Brent Meeker' via Everything List <everyth...@googlegroups.com> wrote:
>> In the final analysis the only way to test a theory is by making objective observations about the way things behave, but consciousness is a subjective phenomenon and that's what causes the problem. The theory that other humans besides me are conscious is perfectly consistent with all observable evidence, but so is the theory that I am the only conscious being in the universe,
> But that's inconsistent with the theory that consciousness is instantiated by physical processes in the brain.No it is not. Your brain operates differently than my brain, if it did not then we would be the same person. Only one chunk of matter in the observable universe operates in a johnkclarkien way, and the theory that the johnkclarkien way is the only way consciousness can be produced is perfectly consistent with all observational evidence available to me. And even I am not conscious all the time, not when I'm sleeping or under anesthetic and I almost certainly won't be conscious when I'm dead either.
> And that theory is supported by many observations and experiments on brains and the reports by subjects.
Many? When it comes to consciousness I have one and only one data point to work with, and there are an infinite number of ways to draw a line through a single point.
>> and so is the theory that EVERYTHING is equally conscious, even grains of sand, even atoms, even quarks and electrons. The trouble is ANY consciousness theory will fit the observable facts just fine, and that's why ALL consciousness theories are utterly useless, except for the theory that solipsism is untrue, that one has a use.
> But that's the way all theories are. We provisionally believe the ones that are consistent with the facts
All theories of consciousness fit the facts,
the same can certainly NOT be said of theories of intelligence, that's why consciousness is easy but intelligence is hard.> And the theory of minds, that other people (and animals) are conscious and have an internal narrative, is extremely useful and in fact any human lineage that did not hold that theory has already been eliminated by evolution.I agree, but if consciousness is not the way data feels when it is being processed (which I have a hunch is true even though I will never be able to prove it)
then a non-conscious being could still calculate how its own actions are likely to affect the environment in the future, and part of that environment would be other non-conscious beings, who also calculate what affect their actions will have on the environment in the future. When 3 grains of sand interact in a Newtonian gravitational way, one grain of sand changes the position of the other two grains, and the other two grains change the position of the first grain, however that is not evidence that the 3 grains of sand are conscious. Of course it is not evidence that the 3 grains of sand are not conscious either.
> Did you ever read Julian Jaynes "The Origin of Consciousness in the Breakdown of the Bicameral Mind"?
No I haven't read it, I have heard a little bit about it and my first impression (which I admit may be unfair because as I've said I have not read the entire book so maybe parts of it are good) is that it just proposes yet another theory of consciousness that is no better and no worse than every other rival theory of consciousness.
>> When it comes to consciousness I have one and only one data point to work with, and there are an infinite number of ways to draw a line through a single point.
> Really?
> Do you really reject the theory that other people are conscious in a way similar to you?
> There are certainly similarities of intelligence, including the ways in which we a tricked by illusions and priming by words. I think the "hard problem of consciousness" is made hard by this kind insistence on incorrigible personal subjectivity which if it were applied consistently would make all science impossible:
> "Well I seem to have heard Bob say that the needle pointed to 2.23 but how do I know he meant the same thing that I do when I see the needle point to 2.23."
> With sufficiently bizarre ancillary assumptions. You apparently agree with Bruno that a blow to the head doesn't eliminate consciousness thru a effect on your brain; it's merely a discontinuity in the stream of experiences called "John K Clark" and his brain is merely a construct of this stream.
> Do you think you could be conscious in the way you are without language?
> Empiricists just look of a good enough theory.
And there is no arguing in matters of taste. And because objective empiricism is of no help in understanding the fundamental nature of consciousness, the field has not advanced one nanometer in the last thousand years.
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAJPayv3ZygSj5veYrEfgXSLoji6LhuVAb01i8R7zQ2Hhe4H8wA%40mail.gmail.com.
>> That's the problem, ALL consciousness theories are good enough, they all fit the facts equally well, choosing one is entirely a matter of taste.
So whether consciousness is a function of brain processes or is immortal soul stuff are equally good theories? Both consistent with the fact that alcohol affects consciousness...assuming it affects soul stuff?
> If tomorrow I came up with a theory and implemented it with a machine that could scan any brain at any moment and tell me what that brain was consciously thinking...an effective theory of consciousness...then
> people like Chalmers would still whine, "But what is it fundamentally?"
If tomorrow I came up with a theory and implemented it with a machine that could scan any brain at any moment and tell me what that brain was consciously thinking...an effective theory of consciousness...then
Then I would ask, how do you know that your machine accurately described what I was consciously feeling?
people like Chalmers would still whine, "But what is it fundamentally?"--
Brent
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAJPayv3ZygSj5veYrEfgXSLoji6LhuVAb01i8R7zQ2Hhe4H8wA%40mail.gmail.com.
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/996098bd-0740-4444-5632-60180466f0dd%40verizon.net.
If tomorrow I came up with a theory and implemented it with a machine that could scan any brain at any moment and tell me what that brain was consciously thinking...an effective theory of consciousness...then
people like Chalmers would still whine, "But what is it fundamentally?"--
Brent
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAJPayv3ZygSj5veYrEfgXSLoji6LhuVAb01i8R7zQ2Hhe4H8wA%40mail.gmail.com.
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/996098bd-0740-4444-5632-60180466f0dd%40verizon.net.
On Mon, Jul 26, 2021 at 4:17 PM 'Brent Meeker' via Everything List <everyth...@googlegroups.com> wrote:
>> That's the problem, ALL consciousness theories are good enough, they all fit the facts equally well, choosing one is entirely a matter of taste.So whether consciousness is a function of brain processes or is immortal soul stuff are equally good theories? Both consistent with the fact that alcohol affects consciousness...assuming it affects soul stuff?
I don't drink but I'm sure alcohol would affect my consciousness and my behavior, and I would be able to prove it affects your behavior too, but I have no way of proving it affects your consciousness, assuming you even have consciousness.
Then I would ask, how do you know the machine is working properly, and how on earth do you read the machine's output?> If tomorrow I came up with a theory and implemented it with a machine that could scan any brain at any moment and tell me what that brain was consciously thinking...an effective theory of consciousness...then
Suppose I'm sad and you put me in the machine and the pointer on the machine's sadness dial moves to the 62.4 mark, does that number enable you to understand what it's like for John K Clark to be sad? I don't think so.
> people like Chalmers would still whine, "But what is it fundamentally?"
Exactly, even if by some miracle you could somehow prove that X caused consciousness they would still not be satisfied, they would demand to know WHY X causes consciousness, and they want to know what caused X.
What do you mean by non effective.
The theory of consciousness (the knowledge that there is a reality) brought by the universal machine, all by itself, is *effective*. It entails immediately the many-worlds appearances (I got it long before I discovered Everett or even QM),
--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/39a2d4f1-cf4d-44b3-a2de-ecb659883dcbn%40googlegroups.com.